Energy technologies are slow and expensive to change. That's why grand predictions of new energy futures are so often wrong.
Anyone purporting seriously to project the global future of energy should first ask themselves this simple question: How much do you think your family’s energy supply will change in the next four decades? (1) Dramatically? (2) Substantially? (3) Hardly at all?
If you’re at least as old as I am (born in 1960), you’ve lived through a good deal of public discussion of energy issues in North America since the mid-1970s, just after the first great oil price shock. Though its roots lay mostly in political causes (a war in the Middle East), the 1973 jump in oil prices spawned a sea change in energy attitudes driven by widespread notions that the world was running out of natural resources.
Many of us were convinced that our energy systems were in for a dramatic shift. If asked then, most of us would probably have said that radical change would occur by the end of the 20th century. We were sure that oil would have become unaffordable. Homes, we thought, would be solar- or garbage-incinerator powered. Cars would be fewer, or electric, and we would likely be driving them much less.
Had you told my generation in 1974 that our sources and uses of energy — and the lifestyles built around them — would be more or less unchanged, not just 20 but 40 years later, those of us who thought we understood energy issues simply would not have believed you.
The lesson is sobering: quick propagation of new technology does not happen in energy supply systems.
The lesson is sobering: quick propagation of new technology does not happen in energy supply systems. Other technologies lend a distorted view of the speed of change, particularly computers and other digital technologies. Our phones, modems, computers and routers get replaced in three- to five-year cycles, while service offerings around them, such as texting, e-mail and social media, seem to be ever-expanding and advancing.
But the energy infrastructure is heavier and more durable (and involves greater safety implications: your computer or phone cannot easily kill you). Light bulbs are about the only thing we can change in three years. Appliances like televisions, fridges and stoves take 10 to 20 years to turn over. Power stations, transmission gear and pipelines last 25 to 50 years; manufacturing equipment 10 to 40; heavy vehicles 10 to 30. Buildings stay in service 40 to 120 years, with much of their embodied infrastructure (elevators, heating and cooling systems) changing just a few times in that period.
With this kind of durability, relatively little of North America, Japan or Europe’s energy-supply hardware is upgraded in a given decade. Add to this the fact that it’s increasingly difficult to get regulatory approval to build infrastructure in our societies (notably pipelines or power lines), and the turnover time is longer still. Yes, you’ll see new technology installed in new neighbourhoods, but not so much in the old ones, at least not for a long time.
So what? Well, two “so whats.” First, if we have a new energy technology that helps to solve an urgent problem such as climate change, adopting that technology on an urgent basis is going to be very, very expensive. If we need to change the system in less than 10 years (and remember, the system hardware has an average lifespan of 30 years) then we have to buy new gear long before we’ve finished paying for the old. That hurts, and in many cases it can’t realistically be financed.
Light bulbs are about the only thing we can change in three years.
Second, note that I said: “if.” Many of our current projections of future energy systems presume the availability of technical solutions that, in many cases, aren’t available yet.
I recently joined a national group that brings together leaders from the electric power industry. These are the executives and engineers who design and build North America’s energy hardware and software, including turbines, wires, transformers and controls. Few are more qualified than these people to talk about what our electric power grid can do and what we can make it do.
In recent conversations I listened to them tiptoe around the gap that exists between rhetoric and reality concerning the future of our electric grid. While they do not discourage efforts to speed up technology innovation (their companies get interesting contracts for pilot projects), they’re aware that many of these innovations are much further out than the general public realizes.
Consider grid-scale energy storage, which is indispensable for intermittent power sources such as wind and solar.
Knowledgeable realists understand that even if wind and solar become economic power sources on the scale necessary to clean up our electricity supply (which isn’t a given), the next big challenge will be to have that power available when it’s needed. At this time, energy storage on a large enough scale doesn’t exist — and won’t exist for at least several decades.
Without storage capacity, intermittent energy sources must be backed up by other sources that can be raised and lowered quickly, such as hydroelectricity or natural gas. If that backup is natural gas, as is increasingly the case, then renewables without storage are not low-carbon and are not an answer to our problem.
There is not yet a good, proven, economic, and widely applicable solution to the problem of grid-scale energy storage. Yes, government funds are being invested in storage technologies. Yes, there are pilot projects. But these shouldn’t be mistaken for signs of an imminent shift in our power system. Ninety-eight percent of installed energy storage capacity in the world today is a tried-and-true technology: pumped hydro. The other 2 percent (including flywheels, batteries and compressed air) are at various stages of development. In all cases, there are significant losses in storing and retrieving the power, with the result that these storage methods are useful only in niche applications where high costs are tolerable.
Another innovation with some distance to go is a workable system whereby those energy customers able to generate small amounts of power from time to time can sell it to the grid (or alternatively, sell stored power, such as from their parked electric vehicles). Considered one of the pillars of the “smart grid,” this is a promising development— unless you are an operator of an actual power grid.
Electrical system operators need to achieve a near-perfect balance between supply and demand. According to a February 2014 article from the Electric Power Research Institute, “The Integrated Grid,” the “electric power grid, especially its distribution systems, was not designed to accommodate a high penetration of distributed energy resources while sustaining high levels of electric quality and reliability.” Bollen and Hassan’s 2011 engineering text, Integration of Distributed Generation in the Power System, tells us that the problems include increased risk of overload and increased losses, increased risk of overvoltages, increased levels of power-quality disturbances, and impacts on power-system stability and operation. If enthusiasts for our smart-grid future dig into such material, they don’t reflect it.
This is not to say that engineers believe these problems are insoluble. They simply believe solving them will take a long time and a lot of investment. It is unlikely that a proven, economic solution will be available in a time frame that lives up to current optimism about smart grids and a future based predominantly on renewable energy.
As in 1974, there are four main barriers standing between us and a dramatically different energy system:
- Engineering feasibility — or the physical problem. Finding new energy solutions is not just a matter of making a public policy decision and pouring millions of dollars into a technology development fund. Many technical problems are simply not amenable to human will.
- Verification — or the standards problem. Performance and safety need to be validated before infrastructure investment can begin. For new technologies, just creating evaluation processes and design standards can be a huge undertaking, and we’re not nearly there yet.
- Infrastructure feasibility — or the time problem. While climate change demands an immediate solution, the scale of our energy system and the rate at which we rebuild it will necessitate an investment time frame in the area of two to four decades — longer if economic growth slows.
- Economic feasibility — or the customer/voter problem. However quickly you and I want a solution (like a mostly renewable power supply) to a big problem (like global warming), quick solutions to big problems are almost always expensive. So while voters generally embrace the solution in the planning stages, they tend to lash out when the bills arrive. The policy may survive or be reversed; the political party may survive or be rejected. But a legacy of large costs is guaranteed.
The engineering reality is that among the low-carbon ways to generate electric power, nonhydro renewables such as wind and solar probably won’t supply more than 5 to 10 percent of our power a decade from now, and 10 to 15 percent in 20 years. Another engineering reality is that the smart grid — so far — is mainly a set of measures to trim consumption and flatten out the peaks and valleys that occur throughout the day. Progress may very well be made on energy storage and the smart grid, but in both cases they’re going to take time.
Let me be clear: I’d like nothing better than a smart grid fuelled by low-emitting generation. Replacing fossil-fuelled vehicles with electric ones would be wonderful for my grandchildren’s ecosphere. I don’t discourage a vision of an energy system that is very different from the one we have today, any more than I underestimate the competence of the engineers I’ve met. I only maintain that realistic time frames need to be considered.
If we agree that global warming is real and the need to decarbonize our energy system is urgent, then we need to institute a lower carbon system on a very large scale as soon as possible. Unfortunately, the list of available ways to do this is very short. The list does not include natural gas, except as a stepping-stone where it is replacing coal or oil (which is the only case where natural gas can be considered lower-carbon). And the list does not include wishful thinking about new technologies we don’t have or big engineering challenges that haven’t been solved. The 60- or 100-year solutions are good and should be pursued. Just don’t be under the illusion that they can help us right now.