In a previous article I considered how the UK might meet its electricity demands after closing its coal power plants. That was a pure replacement problem because there is no new demand growth to meet (indeed demand in the UK has been falling).
In many markets though, demand is rising, presenting a very different challenge for policymakers. The market opportunity for building clean energy isn’t just one of replacing conventional power plants; it’s one of that plus satisfying growing demands. Growth may be enough to provide market opportunity to build both renewable and conventional power supplies.
Let’s take Egypt as an example – after all it was identified in Exxon’s latest “Outlook for Energy” as one of their “10 key growth countries”. As I did with the UK, I’m not going to look at energy overall, but concentrate on electricity.
The aim is not to predict the development of the Egyptian power sector, but simply to highlight some of the key considerations and variables when quantifying future power mix scenarios. In part one we’ll look at the potential scale of demand growth, plus the scale of new power plant deployment required to match demand in energy (GWh) terms. Then in part two we’ll consider whether a suggested power mix also provides reliability; in being able to meet instantaneous peak demands (in GW terms).
The resulting charts come from an energy mix model of my own creation; one which I use primarily as a training tool on courses I deliver.
Part 1: Providing Enough Electricity to Match Growth
Egypt’s electricity generating mix is dominated by gas, plus some oil and hydro, and a tiny bit of wind. That’s charted below (in energy terms).
Figure 1: Egypt’s power generation mix
There’s also an even tinier bit of solar, but that would barely register on the chart above.
Electricity demand is currently around 1700 kWh per capita annually. This is way below both developed markets like the UK and the world average (according to the World Bank stats shown on this next chart).
Figure 2: Electricity consumption per capita (kWh per year)
Unlike the UK though, electricity consumption per capita has been rising; doubling in the past 15 years. So too has population, though at a rate that has been slowing and is now at around 1.6% per year (by comparison, the UK’s is 0.6%).
These factors all feed into an overall growth in electricity demand. Obviously making any sensible “forecast” of the future of these different metrics is fraught with difficulty. Instead let’s consider two scenarios: low growth and high growth, as follows:
- For low growth, let’s assume that demand per capita doubles again, but over the next 40 years rather than 15; and that population growth falls to UK levels in that same period.
- For high growth, let’s assume that demand per capita triples (to a point still below current UK levels); and that population growth falls, but to 1%.
In both cases, I’ve assumed that grid losses reduce from their current level of over 11% down to more like 6% (as typical in markets like the UK or USA). That helps, by reducing the gap between electricity consumption and the amount that needs to be generated before those grid losses.
Here’s what future demand for generated electricity looks like, compared to current generation (i.e. the shaded bars assume that everything current continues generating, or is replaced like for like). Whether growth is “low” or “high”, there’s a significant need for new generation!
Figure 3: Worst case and best case (?) annual electricity demand growth
This could be Egypt or any number of other markets, where economic development, more “western” lifestyles and population growth all combine to produce scary-looking quantities of new power generation to be delivered. In hot countries like Egypt, demand for cooling alone is likely to bring big increases in electrical load.
For now, let’s work with the Low Growth scenario: after all, it’s not unreasonable to suppose that continued increases in efficiency (in various guises: appliances, lighting and so on) can deliver substantial economic development and lifestyle change in combination with much slower growth in electricity consumption. On the other hand there could be factors working the other way, for example new electricity requirements if electric cars become widespread.
The US EIA has a good page summarising the Egyptian energy picture, with the references from where their conclusions come listed at the bottom of the article. They include information on forward plans in Egyptian power generation.
Though seemingly at odds with current global sentiment against this fuel, Egypt has signed deals for the construction of two coal plants of a combined 5.6GW of capacity. They also have a preliminary deal to build 4.8GW of nuclear. Let’s bring coal online within 10 years and nuclear after that (for simplicity, the model I’m using conceptually shows these as smooth additions: in reality obviously they would produce “lumpy” changes).
I’m also going to assume that, over the same period, Egypt decides to phase out its oil-fired generation. It has been rapidly moving from net oil exporter to net importer and its economic growth will certainly involve increasing transport too, further racking up domestic oil demand. So if it can, it would seem wise to avoid burning such a valuable commodity for power too.
As far as I know, there is little scope to build new hydro. However there’s plenty of potential for solar and wind. According to references quoted by the EIA, targets include 3.5GW of solar over ten years and over 7GW of wind sooner than that. I’m going to up these to a suggested 5GW of solar and 10GW of wind over ten years. With reasonable capacity factor assumptions for each generation source the resulting energy generation stack changes as follows, as a result of the capacity additions described so far:
Figure 4: Egypt’s electricity generation mix based on potential (announced) forthcoming deployments
For the first 10 years, notwithstanding the comments previously on the “lumpy” way in which coal will actually add, this at least seems to add up! In practice any lumpiness would almost certainly be smoothed out by variations in gas generation.
So we need to continue building capacity beyond that time horizon. Hopefully much of this can be solar and wind. As with oil, Egypt has been contending with falling gas production and rising consumption, and has started to import. So from an energy security point of view, it’s probably worth avoiding a growth trajectory which makes that balance any worse than it could be.
Now it’s really a case of playing around with scenarios. Below is one which will fill the energy gap. It involves a further 18GW of wind, 18GW of Solar PV and 10GW of Solar CSP. I still needed to find a bit extra towards the end of the model, but wanted to minimise carbon emissions, so I added some more nuclear (maybe by then it’ll be nuclear fusion!).
Figure 5: A low-carbon scenario to meet annual electricity demand in a “low growth” case
As well as meeting electricity generation needs (in energy terms), there are some other results that can be derived from a model like this.
For example: even though more than doubling our generated electricity, the CO2 emissions from power plants actually fall by 4%. It’s possible to factor in assumptions on changing capital costs (e.g. continued reductions in those for solar) and work out how much this will all cost: I come up with a grand capital cost total of $150bn (over the 40 years). Even if gas costs stay the same as today, almost as much will be spent on fuel for those gas power plants – allow the price of gas to double over 40 years and gas fuel costs are nearer to $200bn.
So in summary, it needs lots of large numbers – for capacity, for capital spend and for conventional fuels – to cope with even our low growth scenario. There are obviously many alternative scenarios that could be suggested, with different balances of capital vs. fuel costs and emissions impacts. All of them would involve big numbers. That’s the challenge of growth!
However before we go off and start building, we need to remember that our new power system isn’t just about delivering enough GWh each year. It’s also about delivering them at the right time.
I’ll discuss that in Part 2.