80% Renewables by 2050? Show me a realistic plan!

Posted in Uncategorized at 3:56 pm by Administrator

Many green energy advocates have latched onto a report released by NREL on March 26, 2013 that concludes that 80% of U.S. Electrical generation could come from renewables in 2050 based upon the use of technologies that are available today.

I have the greatest respect for NREL and generally take everything the organization produces as the closest thing to the truth that is available when it comes to talking about renewables. But in this case I would have to say that there is a major element of common sense missing from this report.

NREL have brilliant scientists that utilize sophisticated computer models. I don't dispute the effort that has gone into making these models as accurate as possible but I would say, having worked with similar models in the oil and gas exploration industry, that sometimes the simple facts of life are lost in the 17th decimal place.

So let's keep things simple.

There are two forms of renewable electricity generation that have been deployed en masse in the United States and around the world during the past 20 years; Photo-Voltaic (PV) solar and wind turbines. There are other technologies such as CSP solar, geothermal, biomass, small-scale hydro etc.; but when you look at the total activity in the sector more than 95% is PV solar and wind. Without very major shifts in public policy the safe assumption is that these technologies will continue to dominate.

Electricity demand curves in North America all show a similar pattern; peak demand starts in the late afternoon and keeps going up into the late evening as people get home from work, cook their suppers, turn on lights and turn up the air conditioning in the summer or the heating in the winter. Commercial and industrial demand has somewhat different patterns and in some cases more flexibility but the bottom line is that demand in the evening is high.

PV solar has no ability to store energy using any technology available today and therefore it fades dramatically in the late afternoon. By 5:30 pm PV solar is basically out of the picture.

That leaves us with wind. To understand how feasible reliance upon wind is it is useful to examine what has happened in the United States over the past 10-15 years.

Supported by Federal Production Tax Credits, Renewable Portfolio Standards, and direct subsidies, deployment of wind generation has grown dramatically during that period of time. By the end of 2012 the "nameplate" capacity (the maximum amount of generation from a wind farm under good operating conditions) has reached 60 GW.

The average demand across the United States in 2012 was about 428 GW with peak summer demand topping out at around 750 GW. Based upon the annual average, the available wind generation had the theoretical potential to meet about 60/428 = 14% of demand.

The actual contribution of wind generation in 2012 was about 3.5%. Therefore the effective capacity rate for installed wind farms in 2012 was about 3.5/14= 25% – a figure which is very much in line with effective wind capacity reported from other jurisdictions around the world.

So what would it realistically take to use wind generation to meet peak demand in 2050 using the assumptions that NREL used?

First, NREL allowed for 20% non-renewable generation so the peak summer demand then becomes 600 GW.

Next, let’s assume that a very aggressive Demand Response program has been successfully implemented throughout the U.S. allowing for a reduction of another 25% of demand at peak times (note that post-Fukushima Japan which had lost 28% of its generating capacity overnight was only able to cut peak demand by 15% and that was only in the first year). That brings peak demand down to 450 GW.

Finally, let's assume that aggressive development of biomass, small hydro, and geothermal was able to bring 50 GW of reliable renewable generation on-line (that’s far more than has ever been developed to date). That reduces peak demand to 400 GW.

Based upon an average 25% effective capacity it follows that at night in the summer it would be necessary to have something like 1,600 GW of installed wind capacity to reliably meet demand. At a construction cost of about $2/watt it would cost $3.2 trillion to build this capacity. It should be noted that the wind generation industry would have to more than triple the record 2012 installations and maintain that pace for the next 37 years in order to have an installed capacity of 1,600 GW by 2050.

This might actually be a "best case" scenario because it assumes no increase in the demand for electricity for the next 37 years. Conservation and energy efficiency might make that plausible except for one thing. It is highly probable that the transition to electrically powered vehicles would have been largely completed by that time. The amount of energy consumed by vehicles in the United States today is roughly equal to the total amount of electrical energy produced today. So that implies that the electricity demand could in fact double by 2050.

But that is only the beginning. Wind energy advocates make the argument that it's "always windy somewhere" so that the only thing needed to solve the variability problem is the ability to transport electricity from a windy area to a calm area. The NREL report includes an interactive map showing energy flows across dozens of very high capacity transmission corridors extending tens of thousands of miles (one frame from the flash video is shown below).

The map below shows wind potential as well as the average demand and existing wind nameplate capacity by state. The need to move energy from high wind potential to high demand areas thousands of miles away is obvious.

So how feasible is it to imagine this physical supergrid being built?; Here again, it is useful to look at recent experience.

In 2005 Texas authorized the construction of more than 3,500 miles of high voltage transmission lines to carry west Texas wind energy to the urban centers in central and south Texas. Having expedited the various approval processes this project is nearing completion nearly 8 years after it was started. Cost over-runs have driven the project cost up to $6.8 Billion.

In another recent example the state of New York has authorized $2.2 Billion to build a 1 GW transmission line 300 miles from the Quebec border to the city of New York. That project was announced in 2010 and is expected to be completed in 2017 if it can overcome several legal challenges being brought forward by environmental groups.

Based upon these two recent examples the construction costs for high capacity transmission lines range from $2 million/mile (Texas) to $6.7 million/mile (New York). Given that many of the envisioned cross-continental lines would traverse mountains and large rivers whereas the Texas lines crossed mostly flatlands it could be expected that the cost of building the physical supergrid would be something between these two amounts.

Assuming 100,000 miles of new transmission lines at a cost of $4 million/mile it would cost $400 Billion to complete the supergrid.

The bigger problem is that each leg of this project would have to go through lengthy reviews and would probably face citizen protests and lawsuits so that it would be safe to assume that it would take 6-10 years to build any of the legs. It is very hard to imagine that the majority of this grid could actually be built in the next 37 years no matter how hard we tried.

So where does that leave us? If we managed to triple the rate of wind installations and solve the engineering, environmental, and legal problems associated with building out a supergrid and find about $3.6 trillion to pay for it all we could possibly have an economy driven by wind energy including during peak demand nights in the summertime. (By the way, if this is actually the strategy we want to pursue then there is no point installing any PV solar and certainly no justification for subsidizing it. If wind has to carry the load in the evening and at night it will have more than enough capacity to deal with daytime demand so that all of the PV solar would be surplus.)

My common sense assessment of this scenario is simple. It will not happen.

Does that mean that I don't believe that we can transition to renewables by 2050?; Not at all. In fact I think we can transition to 100% renewables by 2050. But this goal will not be achieved by a simplistic approach whereby wind without storage is doing most of the heavy lifting.

Storage is key and that will not be easy to develop but I firmly believe that if the International community dedicated $100 billion to that problem over the next 10 years it would get solved.

Concentrated Solar Power with Thermal Energy Storage, especially if used only at night, would provide cost-effective base-load capacity that could compliment PV solar. There are other public policy initiatives that are needed, all of which are outlined in my "Sustainable Energy Manifesto".

The biggest barrier preventing this transition is complacency and a belief that we are already on the right path. We are not. Major changes are required that will require coordination and cooperation between governments on a scale never seen outside of wartime.

We can make the changes needed. We just have to understand that these changes are absolutely necessary, will not be easy, will take time and money, and will require some sacrifices in terms of convenience and, at times, giving up some creature comforts. Are we ready to accept those realities?

4,389 total views, no views today

1 Comment »

  1. peter@homerenergy.com said,

    April 29, 2013 at 2:36 pm

    Like you, I get very frustrated by renewable proponents that claim we can get 100% of our electric power from solar and wind. Fortunately, NREL was only claiming 80%. This is a distraction from the practical measures we should be taking to get much higher contributions from these technologies.

    Unfortunately, Your focus on peak demand is also a bit distracting.
    In systems with high contributions from renewable energy, the peak shifts to periods with low resource. The optimization of resources to achieve high renewable penetration is too complicated for “back of the envelope” calculations. That is why we created the HOMER software (www.homerenergy.com), which has been used by over 88,000 planners in 193 countries, to look at how to reliably get 50-80% of their energy from PV and wind. In the past, this was only practical for very small systems, but more and more islands that are dependent on expensive diesel fuel are quickly increasing their use of renewables. These are still small systems compared to North America, but the lessons are instructive.

    The minimum load is the more important metric. Until you have saturated the minimum load, it is relatively straightforward to add more PV and wind. You have to worry about ramp rates and minimum load constraints on the units providing spinning reserve, but that can be managed with more reliance on fast-start units and a few minutes of storage or load management. These “smart grid” techniques are more easily developed on small systems, but will eventually be useful on large grids. Maybe that only gets you to 50% in systems with mediocre resources, but thermal storage, such as cooling buildings with ice made when there is excess energy, can get you another 10-20%. Electric vehicles actually make high contributions from renewables more practical, rather than less as you believe. The key is having a flexible system. Electric vehicles, especially plug-in hybrids, give the system a lot of flexibility about when to meet the load and how to provide spinning reserves.

    I totally agree with your skepticism about the practicality of quickly building the amount of large new transmission capacity. That will limit the amount of wind from the Great Plains, so PV has more of a role to play than you seem to believe.

    Current technology is more than adequate to get to ~50% (or 70-80% in places with an excellent wind resource) wherever smart grid technologies and strategic use of short-term storage can be deployed. For now, that is on smaller grids that do not have the regulatory and security obstacles of the large grids. It will take many years simply to build that many wind turbines and PV panels. That should give the politicians time to get their act together.

Leave a Comment

You must be logged in to post a comment.