Extreme weather from climate change and growing urbanization are making cities more vulnerable to loss of electric power and damage to energy infrastructure. Policy makers and users of critical infrastructure services are searching for solutions that increase the resiliency of energy systems but are closely tied to other goals, such as sustainability and affordability. Creating Resilient-Sustainable Infrastructure Solutions, or “RSI solutions,” will require technical solutions, including: intelligence, redundancy, and coupling and decoupling within networks. For example, predictive tools can be used to better anticipate storms, advanced metering can pinpoint outages in real time, and in some cases, social media is emerging as a potential new source for widespread data and communication. Physical changes to the construct of systems will also be a crucial part of solutions. System redundancy is a traditional form of increasing security by allowing electricity many paths to flow over if one is obstructed. Coupling joins systems together, instead of building redundancies, so that the resulting larger structure expands optionality by making more generation units available or by expanding the variety of fuel types that can be accessed to generate power. While this diversity can create security, being able to “island” a subset of this system, or “decouple”, during a major storm also increases resiliency as those sections can now shield themselves from cascading failures. Finally, governments and planning entities must be involved in these efforts and create awareness of RSI solutions to truly move resiliency to the forefront. Coordination is needed across all levels of government and should engage the private sector. These entities should ensure that disaster planning is collaborative, goes beyond physical solutions and incorporates new types of intelligence, stimulates investment in RSI solutions, and drives research into new RSI solutions as well. Governments and planning entities can achieve both resiliency and sustainability using a portfolio of the approaches described and should continue to seek innovative solutions to meet the demands of an increasingly urbanized world facing growing global challenges of climate change.
The convergence of three trends – Climate change, which is causing more extreme weather, urbanization, which is creating a larger human built-environment across which damage can occur, and the growing importance of electricity in supporting modern economic systems – pose unprecedented risks for urban energy systems.
Investment and policy should advance technologies that are capable of improving resilience and sustainability, creating Resilient-Sustainable Infrastructure (RSI).
One promising solution is the enhancement of resilience through better intelligence. Tools include information technology that harnesses big data, use of advanced analytics, and systems with more sophisticated monitoring and automation.
Physical and operational changes to systems are needed in the form of system redundancy, coupling, and decoupling capabilities.
Governments and planning entities are a necessary element of the solution and should pursue improved disaster planning, policies that incentivize investment in resiliency, and increased research and innovation.
Each year, billions of dollars in energy infrastructure are damaged or destroyed as a result of natural disasters, causing significant social and economic disruptions. Climate change and urbanization, especially the growth of megacities, are amplifying these threats, and the frequency and costs of disasters are rising. However, simply restoring infrastructure systems of the past may be ill-advised. Investing in resilient energy systems would enable local economies to better adapt to sudden shocks such as earthquakes and extreme weather events. An increasingly pressing question facing policy makers and users of critical infrastructure services: How should the need for more resilient energy systems be factored into energy policy and aligned with other goals, such as sustainability and affordability?
Our Urban Planet: Concentrating Risk
Once limited to the developed world, urbanization has become a dominant worldwide force. Many of the world’s largest urban areas are becoming ‘megacities’—fast-growing megalopoli with populations of five million or more.1 These large urban centers will create and consume a large share of the world’s economic output, and they will comprise most of human-created physical assets.
At the same time, the world is electrifying at a rapid rate. In the past four decades the installed base of power generation world-wide has grown almost 800 percent, which is 5.8 times more than world population growth and 4.4 times more than the global economic growth since 1970.2 While this growth has been an important enabling force behind the growth of cities around the world, the size of electricity networks has exposed more of this foundational infrastructure to cyclones, earthquakes, droughts and floods.
The unprecedented size of cities and the energy networks that have grown to support them magnifies the risks of infrastructure failures; not only are more people directly at risk, but the repercussions may affect interconnected systems.3 Flooding and storm surges pose a particularly severe threat because over half the world’s population lives within 60 km of a coastline,4 and another billion live within the path of what used to be referred to as the 100-year flood. By 2015, 21 of the 33 cities over ten million will be on coastlines, all but six in the developing world.5 Since the 1960s, climate change and urbanization have more than tripled the number of reported disasters, to about 320 to 350 large disasters per year.6
When disasters strike, urban energy systems are especially vulnerable because these systems are not only concentrated geographically but also deeply interconnected. Large-scale failures in one part of the system during a disaster can cascade into failures in all systems. For example, power outages caused by Hurricane Katrina and Hurricane Sandy contaminated local water supplies, stopped phone service, and disrupted the availability of gasoline and diesel, hampering the ability to move goods and services.7
Of course, disruptions that happen rapidly are not the only source of concern. Gradual climate changes also pose risks to energy networks. Decreasing water availability can reduce the availability of hydroelectricity and even cause power plants that depend upon water-based cooling systems to shut down.
Not all countries face the same kinds of risk. This variation is revealed when the global installed generating capacity of economies around the world is mapped against the measures of natural hazard risks.8 As shown in Figure 1, approximately 8 percent of world generation capacity, can be found in Asia and Latin America. This group of countries supports 15.3 percent of the world’s population and just under 10 percent of economic output. In particular, Japan scores relatively low in vulnerability due to strong domestic institutions, but this is offset by the country’s very high exposure to hazards, particularly earthquakes and cyclones. The highest-risk emerging markets include the Philippines, Vietnam, Indonesia, Chile and Guatemala.
A higher proportion of the world’s power system falls into the second quartile of medium-high risk. This group represents 28 percent of the global installed base of electric power generation. Many highly populated and rapidly growing economies fall into this category, including China, India, Nigeria, and Pakistan.
Resilient and Sustainable Infrastructure
The impact of natural disasters on energy infrastructure has drawn attention to the need to build more resilient energy systems. A growing number of countries are incorporating resilience into their national strategies. For example, the UK and Australia have developed comprehensive national strategies and institutional arrangements to introduce resilience into their critical infrastructure planning process, recognizing the relationship between critical infrastructure resilience, disaster resilience, and community resilience.9 In the United States, the Critical Infrastructure Task Force of the Homeland Security Advisory Council argued that the government’s critical infrastructure policies were focused too much on protecting assets from terrorist attacks and not focused enough on improving the resilience of assets against a variety of threats.10
These strategic roadmaps converge on a common view that is more holistic than traditional risk assessments. They view resilience as the ability to endure stresses or shocks, and to recover more quickly after a disruption. This reflects a recognized need to expand from preventive measures such as asset hardening to adaptive measures that minimize the negative consequences of disruptive events.
Thus, an important question arises: how does the need to create more resilient energy systems intersect with the need to establish cleaner, more efficient energy systems? Ideally, investment and policy should advance technologies that are capable of improving resilience and sustainability, creating Resilient-Sustainable Infrastructure (RSI).
Hurricane Sandy demonstrated that green energy systems can fail when they are not sufficiently robust. Many homeowners in the region were shocked when their solar systems became inoperable as long as the electricity grid was down.11 This situation occurs because regulatory code requires grid-connected solar systems to automatically shut down when the power fails, partly to protect linemen working on power restoration unless they have more expensive protection switches.
Available RSI Solutions
Technology can play a role in providing the next generation of RSI energy solutions. Three categories are particularly important today: intelligence, redundancy, and coupling and decoupling within networks. The latter involves how energy assets can gain or lose resilience by being part of larger networks.
Enhancing resilience through better intelligence is one promising approach. This involves using information technology to harness big data, advanced analytics, and more sophisticated monitoring and automation. Examples include the following:
- Anticipation and preparation for major storms has improved, with the help of predictive tools that range from simple storm classification tables to more sophisticated computer models that take into account other system variables like topology, system design and layout, customer density and vegetation. Thanks to the application of supercomputers, radar, and satellites, today’s four-day weather forecast is much more accurate than a two-day estimate was 20 years ago.12 More accurate forecasts provide longer lead times for pre-positioning equipment, warning exposed populations, and other proactive measures.
- Crisis management and recovery has benefited from smart meters and advanced Outage Management Systems (OMS) that automatically detect a fault, isolate the faulted section from the grid, and restore service to unfaulted sections. This can reduce the time and frequency of outages as well as the costs of locating the fault and of manually operating switches. These systems can also improve safety for the public and utility worker since faults, such as downed wires, are cleared quickly and utility workers can efficiently manage their work since they can remotely visualize and control much of the distribution grid.
- Social media can be leveraged to more quickly determine the location and extent of a problem and to communicate with customers. If customers link their Twitter tags to their utility account information or enable their mobile devices’ geo-tagging function, they can reveal their location automatically when they tweet. Automatic systems can analyze clusters of tweets tied to addresses to reveal valuable information about location and extent of an outage.13 This provides a significant amount of data that can be analyzed and visualized by the operators, as well as by maintenance and field crews. Advanced grid analytic and visualization technologies can also incorporate social media.
Redundancy is a time-tested approach to achieving resilience; for example, having multiple production facilities provides flexibility in the event that one facility is shut down. A common way to establish electric power redundancy is to install backup generation. Many facilities, ranging from refineries to hospitals, and data centers, have standby generation systems. In many countries regulations often require certain critical facilities to have backup power systems.
However, redundancy can conflict with sustainability goals, as in the above example of backup diesel generators. Thus, to the extent possible, redundant systems should have minimal environmental impact; for example, gas-fired backup generation is cleaner than diesel systems. Even more desirable are systems based on renewable energy. In early 2013, the U.S. Environmental Protection Agency issued new emission rules for diesel generators in order to cut costs and reduce air pollution.14 Under these rules, emergency diesel generators can only operate for up to 100 hours per year, thus inviting the design of redundancy protection with cleaner, distributed power generation technologies.
Network Coupling and Decoupling
Network coupling is a key feature of most energy infrastructure systems. In the case of electric power, deepening the central grid network can enhance resilience by expanding the number and type of generation units available for dispatch. This is particularly valuable in the case of electricity, which cannot be stored in large quantities and requires matching of supply with demand. However, when a network fails, users may need the ability to ‘decouple’ and operate without the support of the central system. This is particularly true for critical facilities such as hospitals, water treatment facilities and communication systems.
One example of the benefits of network coupling is the Central American Electrical Interconnection System. Completed in 2012, it couples the power grids of seven countries, and consists of over 4,500 towers and 25 substations. The project involved establishing the necessary legal, institutional and technical resources to facilitate private sector investment in power generation and to establish a transmission infrastructure to allow power exchanges among participants. In addition to helping lower electricity costs and reduce the region’s heavy dependence on imported fuel oil and diesel, the interconnection and the ability to more effectively pool power generation resources also provide greater resilience to weather events and other natural hazards. Central America is vulnerable to a wide range of natural disasters including droughts, earthquakes, floods, and hurricanes.15
Another example of the benefits of network decoupling is Combined Heat and Power (CHP) systems, which utilize reciprocating engines or turbines to enable decoupling from the central grid. Extra equipment, and therefore extra expense, is needed to enable a CHP system to operate in both grid-parallel and stand-alone mode. Just as an automobile needs a battery to help start the engine, a CHP system needs a battery pack to help re-start during a grid outage. There are a small, but growing number of examples of where CHP systems have demonstrated the ability to deliver critical services during natural disasters.16
Advancing More Resilient Urban Energy Systems
Governments and planners across the world are acutely aware of the challenges they face in providing urban security and sustainability. Most developed and many developing nations have undertaken extensive disaster planning and have elaborate response systems in place. A burgeoning global effort to share best practices has emerged around the four key elements of disaster management: mitigation, preparedness, response, and recovery.17 However, more needs to be done. Mitigation planning often overlooks new strategies for improving energy system resilience by relying on a combination of introducing greater intelligence, innovative redundancy, and the ability to couple and decouple from the central grid system. For example, the official U.S. Federal Energy Management Agency (FEMA) manual on mitigation of urban hazards does not mention energy system resilience.18 Similarly, a recent World Bank report noted that “the energy sector is under-represented in both peer-reviewed literature on adaptation and in related investments and actions.”19
Rather than taking a broad view of energy system disaster resilience, most disaster plans focus on hardening the existing large-scale networks: strengthening transmission towers, building seawalls around power plants, reducing the need for power plant cooling water, storing larger amounts of fuel at plants, and other measures that mainly treat the large-scale system. But, urban energy resilience is now more than simply hardening the large-scale utility network. Modern utility resilience will be provided by a combination of local generation and storage, increased redundancy, conventional system hardening and stockpiling, and smart grid functionality.
Several steps can be taken to accelerate progress in this area. For example, national disaster planning authorities should reach out more intensively to utilities and other companies in the distributed energy, storage, and smart grid production chains. A huge amount of investment and analysis is occurring in public and private settings that may inform disaster risk reduction exercises while addressing more traditional reliability and service value issues.
Researchers should begin developing new tools and methods of analysis that optimize the resilience and adaptive capacity of urban utility systems. There are many investment tradeoffs in the creation of urban utility systems: pipes can be larger or stronger; redundancy elements can be inserted; more sensors and controls can be installed; distant sources can be hardened; and/or, local sources can be multiplied. These options can be optimized only within a framework of portfolio analysis, with clearly designated objectives and a deep understanding of the benefit streams flowing from each option. Finally, more is needed to encourage innovative technologies.
More also needs to be done to incentivize technology innovation to support resilience, especially innovations aimed at reducing the cost of resilience. To date, national energy R&D programs have provided little or no dedicated funding to develop new technologies aimed at providing site-specific or even more general system-level resilience. Even fewer efforts are underway to incentivize innovation for technologies that blend resilience with lower carbon and other important environmental objectives. One exception is the State of Connecticut which has allocated funding for a series of advanced microgrid projects aimed at ensuring that critical buildings continue to have power during electrical outages.20
Establishing energy systems that are both sustainable and resilient will require new approaches. It will require anticipating and preparing for disruptions, resisting or mitigating the severity and impact of disruptions, responding in the immediate aftermath, and recovering to a fully functional and less vulnerable condition. It will also require more than simply hardening the energy systems. Both resilience and sustainability can be achieved through a portfolio of technologies including intelligent systems such as advanced grid automation, redundancy through distributed generation and storage, as well as the ability to couple with and decouple from centralized grid systems. While some technologies exist to support these objectives, more awareness-raising and innovation is needed to ensure that resilient and sustainable energy infrastructures can keep up with the growing global challenges of climate change and the shift to an increasingly urbanized world.
Authors: Dr. Peter C. Evans is Vice President of the Center for Global Enterprise. Dr. Peter Fox-Penner is Principal and Director of The Brattle Group and author of Smart Power: Climate Change, the Smart Grid, and the Future of Electric Utilities. The authors would like to extend a special thanks to Heidi Bishop for her contributions to this article.