Today’s smart grids will power the smart cities of tomorrow, but how secure will they be?
What makes today’s power grid “intelligent” is the modernization of the technologies that both provide and support power distribution. These technologies use intelligent data analysis and communications, via a two-way, automated communications network to analyze what is going on within the grid.
Information about the activities of both suppliers and consumers is collected, analyzed, and acted upon. The goal is to analyze and optimize the many metrics – efficiency, reliability, economics, and sustainability of the power distribution infrastructure.
Smart grids today
Modern smart grids vary from location to location and country to country. In the United States, Texas, Maryland, Arizona, and California are further along the smart grid track that others. However, as a whole, North America is behind some of the other major countries of the world. The Asia-pacific region is the leader, Europe is next, followed by North America.
But this is just a snapshot of the pace of adoption. The higher the cost and the harder it is to get enough power, the more some regions will start feeling the pinch. Nevertheless, it’s becoming clear that smart grids will be the wave of the future. The only variable is when they will get widely implemented. That’s just the first step, too. Smart grids are inextricably linked to smart cities. To have smart cities, especially with the evolving Internet of Things/Internet of Everything, the smart grid will play a very significant role.
The smart grid will consist of modern power distribution and feedback systems via automation, remote control and monitoring and, most importantly, self-healing designs. That will facilitate reliable, secure, and safe integration of the various distributed, and renewable energy resources. All these add up to an energy infrastructure that is more reliable, more sustainable, and far more resilient. Thus, a smart grid sits at the heart of the smart city, which cannot fully exist without it.
As much promise as it holds, there are some dark spots. There is a resistance to move as quickly as needed. “It is one of these industries that has been trying to take advantage of smart technologies for a long time, but always seems to find reasons to delay or look for other solutions that are so low in cost they are impossible to fill,” says Syed Hussain, CTO of Aeris Communications. “That is about to change, but only because the industry finally sees a general recognition that they are going to get left behind in managing the smart city.”
To be effective, those changes have to be made on a grand scale, too. “One of the big concerns is that it has to be viewed as a system-level problem,” said Bernard Murphy, chief technology officer at Atrenta. “You cannot throw intelligence into making individual nodes more secure and expect it to all work. There are too many counteracting forces. It has to be looked at from a much higher level.”
Furthermore, one has to keep in mind that the term smart grid means different things to different people. While the most aggressive adopters are implementing all types of smart interfaces, from smart meter installation to segregation of distribution management components to advanced system analytics, others are just doing smart meter deployments. In addition, much of the technology implementation depends on how far along the cities are in bleeding-edge technologies, such as smart network infrastructures, and low-power platforms.
One piece of good news is that smart grid infrastructure today does not have a wide girth of implementation, which means more time to craft complete solutions. In fact, only about 10% of the available technologies have been implemented across the board, and that there is a tremendous amount of growth potential. New York is a typical example of that, and they face challenges, but is continuing to deploy, successfully, new technologies, despite these challenges.
This bodes well for the semiconductor segment, which will supply much of the intelligence and control as the smart grid proliferates. This is the case for smart cities, as well, because smart infrastructure management requires intelligence at both ends.
A classic example is the work being done by a couple of industry leaders, Intel and Qualcomm. These two rival processor manufacturers are working directly with cities to develop IoE technologies that can support emerging smart city initiatives. One example is the deployment of smart gateways that offer a variety of technology integration. Smart gateways will be pivotal for enabling platforms such as smart buildings, various flavors of wireless technologies, smart lighting and an intelligent vehicular infrastructure. And all will be inextricably woven into the smart grid/smart city framework. New York, for example, is using a private network to install an advanced metering infrastructure (AMI).
Smart grids, smart cities
Perhaps the predominant technology that will tie both smart entities together is wireless. For example various meters such as gas, electric and water are using the cellular infrastructure as the communications medium. “Cellular is everywhere” says Hussain. “The availability of ‘network supply’ is much easier when one know that just about 100% of the population as well as remote power source areas are covered.”
In dense urban areas, mesh networks are finding popularity because many of the suppliers have access to spectrum they own and can build such networks at minimal cost. For applications that do not require high monitoring capabilities (such as clean power businesses), there are implementations of technologies that come under the Low-Power Wide Area Network Alliance (LoRaWAN) specification. These are specifically for low-power sensors, which will proliferate the IoE.
Expect to see this come on line prolifically in the next few years. For the chip industry, this is another huge area for hardware deployments. And there will be hybrid technologies of both, as well. Other technologies may include satellite, fiber, and even copper.
Technology and hardware, especially MCUs, CPUs, FPGAs, DSPs, and simpler devices such as CPLDs, will enable appreciable levels of communications (wired as well as wireless), automation (in the form of self-healing networks), and artificial intelligence to implemented in the power distribution systems of utilities. Moreover, these devices will be found in the receiving end – the smart city – in ever increasing number. Such chips will form the basis upon which smart grids and cities will emerge. Most significantly, they should be the devices that provide security as well.
The anticipated result is that it will improve reliability, performance, security, and efficiency of the smart city power infrastructure. For example, sophisticated levels of intelligence are being designed into substations and field equipment to make better and more accurate analysis of fault isolation, feeder reconfiguration, voltage and reactive power management. This improves fault location and assessment, and restoration times.
Simultaneously, the proliferation and integration of renewable energy sources will require agile and intelligent assessment, and switching. Finally, intelligent power grids will be able to deliver precise amounts of power, exactly when and where needed, via complex feedback systems.
• Enabling real time supply and demand metrics.
• Self-healing from power disturbance events, man-made and natural.
• Providing high-quality power for critical 21st century applications.
• Accommodating various generation and storage options.
• Optimizing assets and operating efficiently.
• Operation resiliency against physical and cyber-attack.
These advanced platforms of integrated information and operations technologies will be the critical element of effective overall design.
Big data and the smart grid
One of the more disruptive platforms that will integrate, deeply, into smart infrastructures is big data. This has real potential to optimize the smart infrastructure. But there is a caveat.
“One of the problems with big data is that there is an implicit assumption made that all data is important, and it should be stored so you can analyze it from all aspects, anytime you want,” says Hussain. “That is a mentality that has proliferated since the inception of the big data metric, but the reality has started to sink in.”
The prevailing opinion among experts is that the vast majority of big data becomes meaningless after a certain point or time. So how does that fit into the smart infrastructure picture? One vector is at the edge. That is where much of the collected data has more value than dense or well covered areas. The fact is that big data in highly concentrated areas tends to be redundant. That is true for the smart infrastructure, as well.
Consider Manhattan, for example. It’s a good bet that many residents have similar power habits. The times and days may vary, but the amount of power consumed by appliances that are installed in the city will, generally, fit into the bell curve model. So collecting volumes of data from each individual source will really yield nothing more than a lot of noise, and it won’t change much over time unless there are significant changes in the collection base.
However, that is not to say it does not have value, especially at the beginning, where it can be used to establish a very accurate general-use model. After that the value is in looking for anomalies in the big data over time. But keeping such data for any extended period likely won’t produce any meaningful information.
“We have to make sure we understand the business purpose to gathering big data, and that is to help make competitive decisions, take some sort of business action, or perhaps assess some sort of cost control,” notes Hussain. In the case of the smart infrastructure, those big data sensor analytics certainly can play a huge role in dynamic allocation of resources at the edge, where density is less and with industries, where shifts can occur very quickly in some cases.
As the smart infrastructure evolves, the massive amount of interconnect increases security leaks by orders of magnitude. It might be easy to secure a power substation, or even a microgrid from access when the direction is delivery only. But smart infrastructures rely on massive amounts of two-way communication, through any number of pathways, each one a potential security hole.
The general consensus is that complete security is impossible, which shouldn’t be a surprise to security experts. However, there seems to be very little interest at present in securing the smart infrastructure. The big issue here is that it will be impossible to assure that only the people who are authorized to get access get access. With traditional grids that is much more realistic than with smart grids because of the two-way communication channels. And with the smart cities connected to the smart grid, there are many more possibilities for breaches.
There are several types of security that can be implemented at various levels to provide a reasonable measure of security until both the grid, and the city are much farther along the development curve. The first is content. That works well for sensitive data, but there isn’t as much of that in the smart grid as there is in other segments, such as finance, medicine, or transportation.
Another is access security. That is a bit more relative. Keeping unauthorized personnel out, whether the access is physical (on site) or virtual (command center computers), is a critical vector to securing the component.
A third area is control security. That involves keeping keys, codes, and passwords protected. If someone can get access to power at a particular location it isn’t nearly as big of an issue as if someone gets control of the delivery of power.
The best way to approach it is to make it as bulletproof as possible, of course, which is at the chip level; something that has been pushed by chip vendors for years. But the utility industry is a private entity with stockholders to account to, which is why there is such an emphasis on cost.
For now, the industry’s prime security concern is having a security breach at a generation facility such as a dam. The results of a compromise at such a facility will have much farther reaching implications than a breach at a distribution facility.
For now, the penetration of smart grids and smart cities in North America at a fairly low level. Much of it revolves around various iterations of smart meters. But there is a considerable amount of traction for getting the advanced components of smart technology in place, both in grids and cities.
Both smart cities, and the smart grid will evolve steadily over the next two decades. When fully online, they will be able to aggregate, integrate, analyze, and act upon, intelligently, data and information across all interconnected elements of both camps. The smart grid will become the supporting platform of an intelligent and reactive smart ecosystem.