On Feb. 9, the Nuclear Regulatory Commission approved the issuing of a combined construction and operating license for the addition of two 1,100 MW nuclear reactors at Southern Company’s Plant Vogtle in Georgia. My first job out of college was doing generation planning studies, and I’m still intrigued by the economic analysis and comparison of different generation technologies, and which generation technologies will be installed in the future. I’m quite aware of the economies of scale that still exist for centralized generation in most cases, even when other factors such as T&D costs are considered.
But being in the distribution field, we tend to be exposed to all the news (and in some cases, the hype) of distributed generation, as well as the falling costs and continued investments in technologies such as PV. In reality, the evaluation of the amounts of different generating technologies that will be installed in the future is a complex task, and is based upon forecasts of future fuel prices, upcoming technology developments, load growth forecasts, impact of efficiency measurements, and regulatory, policy, and geo-political factors. Just forecasting each of these individual factors is quite an involved exercise. It’s not something that can be done in a spreadsheet over a couple of hours.
I decided to take a quick look at the long-term forecast of the amount of different types of generation to be installed in the US over the next twenty or so years. The US Energy Information Administration, which is the statistical and analytical agency in the US Department of Energy, produces an Annual Energy Outlook (AEO) for the US. In late January, they released an Abridged Version of the 2012 AEO. The full AEO for 2012 will be available later this spring.
The chart below shows the forecasted amounts of generation capacity in the AEO reference case. The chart includes both electricity-only and combined heat and power plants whose primary business is to sell electricity, or electricity and heat, to the public, as well as end use generators.
The US presently has about 1,038 GW of generation capacity. The generation additions shown represent an average annual increase of 0.6% per year in the US, with a net increase of 158 GW over the next 25 years. That’s a net increase, including the estimate that oil and natural gas steam plus coal will decrease by about 28 GW in this time frame. The gross increase, or new generation to be built, is 196 GW. The increase in nuclear generation is forecast to be approximately 11 GW, which includes capacity uprates of existing units.
What’s left is the amount of new generation to be powered by natural gas and renewables. These total about 185 GW of new generation capacity. Combined cycle, combustion turbine / diesel (which will be largely natural gas), and distributed generation (natural gas) make up about 64% (119 GW) of this amount, with renewables accounting for the other 36% (66 GW).
If you’re work in distribution, you’re concerned about the amount of distributed generation that could occur on your distribution feeders. For the AEO forecast, a logical assumption is that the only generation that might be placed at distribution level voltages are renewables and distributed generation (natural gas). The breakdown in renewables is shown in the chart below.
Wind generation makes the most significant contribution to renewable growth, but due to economies of scale, it is almost always connected at transmission level If it is assumed that 95% of the wind generation, geothermal generation, and hydro is tied in at transmission level (and not distribution level), and if you assume that 100% of the solar thermal and solar photovoltaic is connected at distribution level (which actually likely won’t be the case, because of larger solar farms), and assume that 50% of wood / other biomass is connected at distribution level, approximately 25 GW of renewable will be connected at the distribution level. Adding the 2.8 GW of distributed generation (natural gas) from the first chart, the total connected at distribution level is around 28 GW, which is about 14% the total gross increase of 196 GW.
Here are my thoughts on this:
1. First, my hedge. Forecasts are almost always wrong. I’m not saying the EIA does a bad job – I’m sure they’re very good at what they do. But remember that the AEO reference case is a base case type of forecast. There are a lot of other scenarios, in which the contributing factors will be different, whether they are costs of generation technologies, governmental policy and regulation, rate of economic growth, and other things. Could things turn out substantially different? Absolutely.
2. The continued impact that natural gas will have on our energy supply is forecast to be very significant. Of the 196 GW of new generation to be built, it is forecast that natural gas will account for 119 GW (61%). The recent natural gas price declines may be short term, but forecasts show that over the next few decades, supply will be abundant and prices will not escalate severely. Technology advances and increasing levels of shale gas have greatly increased the amount of natural gas economically recoverable. Should policy changes be enacted to lessen our dependence on natural gas, so we have a more sustainable future in the long run? That would be the subject of another column.
3. Large generation technologies that utilize economies of scale are still dominant. This is evident by the amount of centralized generation to be added (86%) versus the amount of distributed generation (14%). Of course, the incremental costs of T&D, both capital costs and operating costs, still have to be considered. And there will be some exceptions – off-grid applications, for example – where distributed generation will be required. But centralized generation will still be the bulk of new generation capacity added.
4. If only 14% of new capacity additions will be on the distribution system, does this mean you don’t have to worry about distributed generation on your distribution system? It depends. Regional policies such as renewable portfolio standards, tax incentives, rebates, grants, etc. will still lead to significant amounts of distributed generation in some locations. For other locations, the environment won’t be nearly as conducive to strong growth. Just as higher penetrations of distributed generation are already creating issues for some distribution organizations, there will continue to be significant increases in the amounts of distributed generation in some locations. As a whole, solar photovoltatics are forecast to increase seven times in this time period, and they will be largely connected at distribution level. Technologies are certainly changing, and the amount of investment is still substantial. Even though the percentage of new capacity additions that will be connected at distribution level is small compared to centralized generation, they will still have significant impacts on distribution in some locations.
