Grid and infrastructure resilience are increasingly important, while a relatively ‘new concept’ in terms of today’s modern grid, and its dynamic environment.
With the increase in natural disasters, and as the northern hemisphere goes into what is commonly known as ‘storm season’, Smart Energy International spoke with Dr Alexis Kwasinski, Associate Professor at the Department of Electrical and Computer Engineering at the University of Pittsburgh. Kwasinski specialises in grid resilience research in areas prone to natural disasters and extreme weather.
According to Kwasinski, grid resilience research has changed a great deal over the years. When he conducted his fi rst study during Hurricane Katrina, the major hurricane that made landfall on Florida and Louisiana in August 2005, fewer people were talking about resilience.
However, as natural disasters such as hurricanes and earthquakes increase in frequency and severity, the term is attracting more attention around the globe. Natural disaster events allow researchers to conduct analysis and draw conclusions, not only concerning the impact of natural disasters on the grid, but also highlighting how governments and utilities should prioritise funding and manage the resources needed to secure higher levels of resilience.
The Institute of Electrical and Electronics Engineers (IEEE) looks specifi cally at the various measures and concept definitions concerning resilience.
When conducting research into grid performance, metrics are based on four main factors:
- Withstanding capability – how long did it take to break,
- Recoverability – how long was the outage
- Adaptation – applying the lessons learned, and
- Planning and operational capability.
According to Kwasinski, measuring resilience is “actually a lot more complex than it sounds and, of course, there is a lot of discussion around standardisation. Complexity depends on, among others, your people, technology, resource management, restoration crew management, managing the sequence of events and how you train your people.”
The impact of grid design
Historically, resilience was not a requirement when designing power grids.
The more important factors included rapid infrastructure deployment, reliability and securing the funds to make it happen. In fact, Kwasinski explains that older grids were designed with greater focus on ensuring power quality across the board as towns and electrification grew. The result is that modernisation of a larger grid can be an enormous undertaking and part of a long term modernisation programme. However, microgrids allow for a more localised strategy, with a different concept of electrification.
According to Kwasinski, the way that grids are designed can have a significant impact on how the grid responds to disasters. One small component that is down could potentially result in loss of a large portion of the grid functionality. This premise forms the basis for his research. To accurately gauge the resilience of a grid one must account for age, design and the effects of specific parts on the whole.
Research into grid resilience can be challenging when a weather event is more random and localised, such as in the case of a tornado, says Kwasinski. “Localised damage can thwart one’s ability to study the system in its entirety in order to really understand the impact of an event on the grid.
“When looking at the impact, people focus their attention more on larger areas with the most damage. However, where there is less damage, it allows you to see the effects of phenomena on specific, isolated parts of the network. Even a little damage in an isolated area can result in many customers losing power. I need to concentrate on the systemic impact, understanding significant outages from little events. However, whether dealing with main or microgrids, the crux of achieving greater efficiency and resilience is all down to cost.
“How much are you willing to pay to minimise the impact of a natural disaster?” he asks.
Energy storage has a significant impact on infrastructure resilience plans. “Access to storage locally allows you to deal with dependency issues. It’s a buffer on the power grid.”
As more renewables are incorporated onto the grid, as well as with increased use of natural gas, it’s vital to measure dependency versus availability. Using buffers such as storage increases availability and is important for true resilience. Such as in the case of solar, the power must be constant and consistent under all operating conditions. That will require a lot of storage to achieve – and that is, of course, very expensive.
“We are back to the question of how much you are willing to pay for resilience.”
The economics of resilience
Kwasinski refers to Puerto Rico to illustrate how the level of resilience achieved is directly proportional to the fi nancial resources behind modernisation strategies. He explains that Hurricane Maria was not solely responsible for the lack of power after the storm. Up to a decade before Maria, the island experienced an economic downturn.
“The utility was losing load for many years prior to the storm. Less capital investment into maintenance, vegetation management and hurricane preparation put them at risk before the storm was on the horizon. They lacked funds before Maria and now the taxpayer has to cover the cost of the damage and the lack of prior investment.” The opposite was seen in the Fukushima disaster in Japan, where the total cost of the disaster was covered by the government.
Says Kwasinski: “How you plan grid modernisation comes down to economics and how the local community prioritises services. Do the locals need constant power, or is availability of supply balanced with other services and resource distribution? Do they need clean water more than electricity? Economically challenged countries often experience this dilemma.”
It is important for decision-makers to consider the broader perspective, indeed a complex matter. Unfortunately, when it is perceived that the government is taking insufficient action, the community or private sector takes issues into its own hands, especially if the grid can’t meet its power requirements.
Kwasinski explains that this leads to the increased use of microgrids, which upsets the equilibrium of the grid and society at large. “The result is that communities are not contributing to a common goal, where everyone gets access to the same good quality power. This lack of democratisation or inequality can cause distress.” Of course, impoverished communities rely most on the grid and they don’t have the financial resources to help themselves by installing expensive microgrids with storage solutions.
The human element
Ultimately, one of the most overlooked components, vital to the success of any grid resilience strategy, is the human element.
According to Kwasinski, “studies suggest that people usually shift their understanding of risk before and after they experience a disaster. When it comes to extreme weather or natural disasters, they tend to adopt an ‘it won’t happen to me’ bias.”
This, in turn, impacts planning and unfortunately, the plans usually tend to be more reactionary in nature.
In the case of the Japanese tsunami that affected Fukushima, the sea walls were simply tossed around by the raging ocean.
However, says Kwasinski, everyone thought they were prepared and safe, that those walls were strong and resilient. “The impact of the tsunami tells us that although they were well prepared for a less intense event, such preparation was insufficient for a less likely, but more intense one.
Kwasinski asks: “The question of course is, how high do you build your walls? How safe is safe enough?”
“This is the challenge with climate change – there is simply no precedent, which makes people less concerned. It’s safe to say, there isn’t enough being done, especially from a policy perspective. People don’t understand the concept of probability – it is human nature.”
Research to the rescue
The data collected by Kwasinski and his team are included in resilience models and plans, such as plans concerning power and communications best practices. Kwasinski gave an example of plans his team presented to the US Federal Communications Commission. These best practices will be used to advise on how to measure the impact of storms on power grids.
Research findings are also presented to the Defence Threat Reduction Agency, especially focusing on weapons of mass destruction and the threat to physical infrastructure. Models are created with the data and used to generate scenarios, allowing for better planning.
As the grid becomes more complex, with the inclusion of microgrids, storage and a variety of hybrid solutions, as it becomes more intelligent and digitalised, we will need to see better and more advanced controls utilised, cybersecurity measures implemented and modernisation strategies deployed. If this does not happen, grids continue to be vulnerable to acts of terrorism and Mother Nature. Both can be equally devastating.
It is up to governments and utilities to ensure the people, policies and financial resources are aligned to support the modernisation process. Perhaps, we also need to encourage a culture of ‘it can happen to me’ and use that as the reference point for truly future-proofed modernisation strategies?
About the author
Dr Alexis Kwasinski is an Associate Professor at the Swanson School of Engineering at the University of Pittsburgh.
Dr Kwasinski is particularly interested in analyzing the effects of natural disasters on critical power infrastructure, such as communication networks power supply, and studying ways of reducing the vulnerability of these critical power infrastructures to such extreme events.
His background is in telecommunications power systems, with a PhD, Electrical Engineering, University of Illinois at Urbana-Champaign, (2007); MS, Electrical Engineering, University of Illinois at UrbanaChampaign, (2005); Graduate in Engineering Specialist Telecommunications, University of Buenos Aires, (1997); and a BS, Electrical Engineering; Power and Energy systems, Buenos Aires Institute of Technology, (1993). He values research in practical applications based on sound theoretical analysis and experimental verification.