What stands in the way of a rapid energy transition? Cost? Unlikely—witness recent cost reductions for solar, wind, batteries, heat pumps, and electrolysis. Material shortages? Maybe temporarily, but wherever you look there are substitutes for scarce resources: rare-earth-free wind turbines and electric motors, copper for silver in PV, batteries without cobalt. Land? On paper we have plenty of that. What about grid stability? Don’t worry, our best engineers are on it.
If I had to bet, my money would be on low social acceptance for new infrastructure projects as the biggest brake on a fast reduction of emissions in the energy sector. In densely populated parts of the world we have seen local opposition grow to new projects: to onshore wind in England, transmission lines in Germany, utility-scale solar in Spain, railway lines in Italy, offshore wind in the United States, carbon sequestration projects in Norway, and nuclear in Japan. Local opposition groups have networked within countries and across borders to exchange best practice on hindering projects in the streets and in the courts. New approaches are clearly needed to address the concerns of people most affected by new infrastructure.
Tom Brown
What Can We Do?
So what can we engineers and modellers contribute to this discussion? A good starting point would be to acknowledge that low cost and technical elegance are only part of the story: energy is embedded in a social context, where other factors like community involvement, perceptions about scenery, and the stories we tell about energy can also play a role in decision making. Building bridges with social scientists is sorely needed.
Clear communication can certainly help, particularly where there have been misunderstandings about the effects of electromagnetic fields around power lines, low-frequency noise for wind turbines, or the risks of nuclear power. However, we should avoid falling into the trap of assuming that if we explain it clearly, people will accept the message. Local groups have genuine concerns, be it about landscape impacts, property prices, noise, or effects on wildlife, which have to be addressed. Trade-offs with the common good will have to be made.
Here’s where energy system modellers can help: by elucidating alternative infrastructure options that stay within a reasonable budget. Often engineers will only highlight a single cost-optimal solution to an engineering problem, i.e., a solution that obeys the physical constraints of the problem (like energy conservation), while achieving the lowest possible cost. However, this single solution belies the fact that there may be a diversity of alternative solutions that have wildly different configurations (e.g., fewer wind turbines or power lines), while costing only a fraction more. Articulating these alternatives can help to balance the trade-offs outlined above between local concerns and global affordability.
Benefits of Near-Optimal Solutions
The exploration of near-optimal solutions was standardized decades ago in the operations research community and first applied in an energy model in 2010 by Joe DeCarolis. It turns out that real-world energy systems can deviate by up to 23% from the cost-optimum, as was shown in a study for the UK in 2016 by Evelina Trutnevyte. In the last few years, many different studies have highlighted the wide spectrum of solutions close to the optimum. To take an example from a recent paper led by Fabian Neumann in which I was also involved: for a fully renewable European power system, the cost-optimum contains a mix of offshore wind, onshore wind, and solar, balanced by transmission grid expansion and storage in batteries and hydrogen (see Figure 1). But if we allow solutions that cost only 10% more (see the x-axis of Figure 2), then whole new vistas open up: we can avoid grid expansion altogether (relying instead on storage for balancing); we can eliminate solar (by substituting with wind), onshore wind (by substituting with offshore), or offshore wind (by substituting with onshore); and we can easily exclude batteries, but dropping hydrogen is harder. We cannot do all these things at once, but by exploring the extremes of the solution, we can understand how much freedom there is to make trade-offs.
Figure 1: Cost-optimal investment in new transmission and generation for a particular set of model and technology assumptions from Neumann and Brown (2020).
Figure 2: As we change the deviation from the optimum (x-axis) we can force up or down the capacities of the technologies (y-axis) away from their optimal values, thus forming a Pareto front between cost and capacity. (Source: Neumann and Brown (2020)
One further benefit of these techniques is that they can raise the robustness of solutions. A single solution can give a false sense of exactness, since there may be a multiplicity of solutions at similar cost. With near-optimal methods, we can see clearly what is superfluous (e.g., the batteries in Figure 2) and what is required for a low-cost solution. For example, in the top left graphic, you can see that at least 500 GW of wind (either onshore or offshore) is required to remain within 10% of the optimum, even though offshore and onshore can be substituted for one another with little cost impact. What is important is that a substantial fraction of wind is in the system, because in Europe wind is available in winter when demand peaks.
While explorations of near-optimal infrastructure will not solve all problems, they can play a useful role in highlighting trade-offs that can increase the acceptance for the energy transition. A new European-funded project led by Stefan Pfenninger called SEEDS (Stakeholder-Based Environmentally Sustainable and Economically Doable Scenarios for the Energy Transition) will use near-optimal approaches to integrate stakeholder feedback on social and environmental impacts of energy projects. For the success of the energy transition, such approaches are very welcome.
Tom Brown
Technical University of Berlin
Great article. Let’s get started!