All images © NRG Magazine
Dr. Hassan Farhangi is a pioneering Canadian researcher in the field of smart grid development. After accepting his position with British Columbia Institute of Technology as Principal Investigator of Smart Grid research in 2007, he became responsible for creating Canada’s first smart microgrid on its campus. In the following interview, he shares his view on some of the topics in the smart grid domain, including cybersecurity and renewables integration.
NRGM: For how long have you been working with smart grids?
Farhangi: I have been involved in the energy sector for more than twenty years. But this particular activity [smart grid research- ed.] began in 2007. In fact, when we started, the terminology was not even coined yet. We had to use the term intelligent grid. We were aware of the problems that utilities had, but there wasn’t a coherent understanding of the collection of concepts and technologies that now formulate a smart grid. We started by trying to describe what it was, the thing we were working with.
Our target was to put a layer of intelligence over the existing grid. Later on, when our neighbors to the south, the Americans, started talking about this, their Department of Energy coined the term smart grid. We, being a smaller neighbor, followed suit.
NRGM: Personally, which part of smart grids interests you most?
Farhangi: In 2007, we started a program together with BC Hydro and Canadian Government focusing on creating a smart micro-grid on our campus. Now the work is almost done, and the micro-grid is there. Its beauty is in the fact that it is a near-real environment, where you could experiment with different things.
Utilities like to have an impartial party with enough scale to be able to develop and validate solutions, and help them decide which technologies should be adopted. The utilities’ infrastructure is just too sensitive to accommodate unproven technologies. BC Hydro and the Government invested heavily in our campus because that’s where testing, development, validation, and all other possibilities are. We have a scaled-down version of a real grid. It is so large that it is fed by three different utility feeders. When we started looking at the AMI, the Advanced Metering Infrastructure, we implemented all sorts of technologies, and started to compare them. That is the kind of real environment that you can never simulate in a lab.
NRGM: Whom would you say you represent: utilities or home-owners?
Farhangi: I wouldn’t say that I represent the utilities, and I wouldn’t dare to claim that I represent the consumers either. This sector has a large number of stakeholders, and it is not as black and white as we would like to think. We, scientists, are working towards a common goal of bringing efficiency into the system.
NRGM: When you use the word “utilities,” whom exactly do you mean?
Farhangi: In North America, in the context of the energy sector, “utilities” are primarily electric utilities, companies and organizations that do a part of the whole thing – some are on the generation side, some are on the transmission and distribution side. But the collective terminology to describe them all is utilities.
In Canada, I work with utilities that have sponsored and supported our work, such as BC Hydro, Hydro Quebec, Hydro one and Manitoba Hydro. These are the largest hydros in Canada.
By the way, in Canada the word hydro was coined to describe the utilities because we have an enormous amount of hydro resources. In BC, for example, close to 98% of power generation comes from hydro resources, the cleanest kind of energy one can imagine.
“WE, SCIENTISTS ARE WORKING TOWARDS A COMMON GOAL OF BRINING EFFICIENCY INTO THE SYSTEM”
NRGM: What is your personal definition of a smart grid?
Farhangi: A grid, as we’ve known it for the last twenty years, is a marvel of the engineering world. If you look at the constituent parts of this grid, there is a lot of intelligence already built into the upper part of the system, in production and transmission.
However, on the distribution side, there has been almost no investment for the past hundred years. What we call a smart grid is an attempt to bring this lacking part up to par.
Look at the utilities flow today. We are generating an enormous amount of energy, blindly pushing it down the line, hoping that everyone on the receiving end is able to use it. While the energy is being generated all the time, households do not need the peak load 24/7.
In short, what a smart grid is trying to do is create a paradigm shift. The system today is basically load-following. You have variable power generation which follows the load, but we want to have a system that is generation-following. Meaning that we want to know at any given point how much power we can safely produce, and control our loads accordingly, matching the demand and supply.
“A GRID, AS WE’VE KNOWN IT FOR THE LAST TWENTY YEARS, IS A MARVEL OF THE ENGINEERING WORLD”
NRGM: In your scientific works you use the term ‘layer of intelligence.’
Farhangi: As power systems engineers, we know that the system is pretty good, it is running and it is stable. However, the amount of investment required is daunting. Imagine if you need to build a car, while at the same time continue to drive at 100 km/h. That is roughly the analogy. In order to do what is required, you need to be extremely strategic and add the intelligence to the system, layer by layer, when and if required. For instance, you can start with installing smart meters to measure the load. Then, you can use the data to optimize energy delivery. When that is done, you might think about accessing other sources of energy. You need to go according to a very well-defined roadmap, and gradually add layers of intelligence over the system.
NRGM: What is the first layer that needs to be there?
Farhangi: Definitely the AMI, the Advanced Metering Infrastructure, is the very first layer, but one of the major issues that we have seen in North America, was that in the haste of making this first step, a lot of utilities missed other important aspects, like consumers. Missing out on the latter created a lot of backlash in Canada and in the US, which prevented the system from moving forward.
NRGM: Overall, how is Canada performing on making the system intelligent?
Farhangi: Very well. Our biggest utilities, BC Hydro and Hydro Quebec, are known in the world for being at the forefront of the innovation and development of the smart grid. They have been the early adopters of a large number of technologies, investing a lot into R&D. There is also an effective collaboration between the public and the private sector in conducting the research. Both utilities and the government invested heavily into our research. Since 2007, BCIT alone has received roughly 20 million dollars of investment for developing the smart grid, integrating different solutions and validating technologies. Our government has been extremely supportive not only investing in R&D, but also working with utilities and the private sector to promote the infusion of a smart grid into Canada’s grid.
NRGM: Speaking about the stakeholders, who has the most responsibility in propelling the system forward?
Farhangi: It’s a partnership. Every stakeholder has a role to play in this environment. From a financial point of view, it’s the utilities that require that level of investment and overhaul to be realized because their primary business is at stake. On the other side, there are national governments who would like to have a stable national grid that contributes to the growth of the economy. These two should work together, helping to bring in the third stakeholder, the consumers. A smart grid can never materialize without the active, direct involvement of the consumers. Unfortunately, as a practitioner in the field, I don’t see that level of involvement materializing any time soon. I would not put the blame on anybody’s doorstep, but all of us, including governments, scientists, and utilities, have failed to understand how important this group of stakeholders is.
NRGM: Who controls the smart grid?
Farhangi: If you look at the smart grid as an abstract technology, then no one. A smart grid is a collection of concepts and technologies that need to work together for certain applications. Mind you, different countries look at the smart grid differently. For example, the Americans tend to focus on expanding the generation side of the system. In Canada, we have a different approach, our primary focus is conservation. Every country and every utility have their own set of priorities. But overall, at this point in time, utilities are in the driver’s seat. They are the ones who determine the pace of investment, at least looking at the surface of things.
NRGM: Let’s talk about cybersecurity and smart grids.
Farhangi: A lot of issues we discussed fade in comparison to the problems we are facing in the cybersecurity area. In the haste of addressing stability issues and trying to bring efficiency in line, I think we forgot that whatever technology we are bringing into the system may not be suitable from a security point of view.
Take communication, for example. The communications system is designed to transport data from point A to point B. When we talk on the phone and there are sound problems, I can ask you to repeat what you said, but this kind of technology does not have the level of availability and reliability that is required for the utilities sector.
Yet we have taken technologies from other sectors, technologies that have their roots elsewhere, into our ‘bag of tricks’. We brought them into the utilities sector not realizing that most of these technologies need to be redone, repurposed and optimized for a new application. Some of these technologies, unfortunately, have security holes associated with them. Because of that, the more we add to our existing system, the more functionality, devices and subsystems, the more vulnerable the system gets.
“SOME TECHNOLOGIES WE BROUGHT INTO THE UTILITIES SECTOR HAVE SECURITY HOLES IN THEM.
THE MORE WE ADD TO THE EXISTING SYSTEM, THE MORE VULNERABLE IT GETS”
NRGM: So the intelligence of the system is its own trap? The more intelligent it becomes, the more vulnerable?
Farhangi: Yes, exactly. The way we have discussed intelligence, our understanding of intelligence, is the visibility, command and control of the system. Command and control by whom? By an authorized person. But what happens if an unauthorized person with mal-intentions enters into the system and gets into the driver’s seat? If you have a meter on your wall, that meter is an entry point into the larger system. That meter is not secure. If you allow the 14-year old kid to sit beside the meter, get into the ZigBee radio and position himself as an installer, and start changing the stuff, there would be chaos.
NRGM: Technically, how easy is it? How easy is it for anyone, say, even me, to compromise the system?
Farhangi: Not easy. Up until now we have had intrusions into the system primarily through networked-devices, new devices that are going into the field, e.g. relays, communication systems, meters, PMUs, you name it. From those devices there have been intrusions. Up until now, the intrusions have been made by hobbyists who do this for fun. Some find it amusing to sit next to a substation and attempt to get control of these new devices and play with them.
NRGM: But that’s the common intention for any hacker.
Farhangi: Yes, but the situation can get a lot more serious if you realize that there is a potential for well-organized, highly-structured, focused intrusions into the system from beyond the country’s borders. Those foreign entities come in and take a look inside, like they would look into your house – what kind of furniture do you have? What kind of a system do you have? Then they just leave without touching anything. It is scary.
There is an American cybersecurity consultancy called Mandiant, you can see their reports online. Mandiant reported more than a few thousand intrusions into the US infrastructure, and some into power plants, by parties from the other side of the Pacific, who came in and looked at stuff. Their conclusion was that the primary interest of such attacks was data. There were a lot of IPs stolen. So, industrial espionage was the primary intent.
Just think about it. If the defenses that we have in our systems could be compromised at this level, then what would the next stage be? A coordinated attack on our power plants and on our transportation system? On our airports? They can create chaos, and that is a fact.
We have to be very strategic and very careful about what we do, and realize the fact that security should not be an afterthought. You cannot install something and commission the project, and then get back to it and think – is that secure enough? By then it is too late. Security has to be part and parcel of the design of anything that goes into the smart grid.
“FOREIGN ENTITIES COME AND LOOK INSIDE INTO YOUR SYSTEM.
THEN THEY JUST LEAVE WITHOUT TOUCHING ANYTHING. IT IS SCARY”
NRGM: Which defense options are there?
Farhangi: Plenty. You could look at the wide variety of the attacks, and for each attack, the defense is different. Some attacks are designed to get over a firewall and try and compromise the system, try and get into the subsystem and change the configurations. So there are defenses against that. Some attacks could be supplying the system with wrong information. Imagine a smart grid as a huge finite state machine. A state machine is a machine that has legal states. The transition between each state is based on getting authorized and valid stimuli. In this kind of attack, you don’t need to get inside the system to disrupt it, you just feed it the wrong information.
We actually tested one of those cases in our campus. On our campus we have an intelligent home, a three-story home built as a Net-Zero Home. We have loaded this home with all kind of gadgets. One of the issues we were looking at, something that utilities wanted us to experiment with, were pricing signals. If you have intelligent appliances in your home, such as an intelligent dish washer, you could program it – tell it to wash something at a cost no more than 1.20 dollars. We were sending pricing signals from our servers to the dish washer and the dish washer was waiting for the pricing signal. The minute the cost of electricity was low enough, it started washing.
What would happen if we started manipulating these prices and signals?
What if all the appliances are manipulated to come online and start working when they are not expected to putting a huge load on the system and compromising it? This is so easy. Imagine if you have a huge environment, maybe a few cities in The Netherlands, and during the peak hour the utility sends the pricing signal to all of the homes, telling them to not start such devices because the grid is reaching an unstable status. If someone comes and negates that signal, it will collapse the system. By allowing wrong information to be fed into the system, you make it collapse. There is no need to break into it.
Organizations with mal-intentions are hard at work. They have groups of highly intelligent engineers who are doing nothing but thinking of a way of attacking a system of this magnitude. We have to put in an equal, if not bigger, amount of resources to stay a few steps ahead of them.
“ORGANIZATIONS WITH MAL-INTENTIONS ARE HARD AT WORK.
WE HAVE TO PUT AN EQUAL AMOUNT OF RESOURCES TO STAY A FEW STEPS AHEAD”
NRGM: The more the system allows communication, the more it allows for sabotage?
Farhangi: Only the non-secure communications are a problem. We have no other choice but to strengthen the system. Everyone involved acknowledges the fact that this is exactly where we are heading, and we need to do this. At the same time, let’s not have the tunnel vision. If you are running at 150 km/h, you better have a 360 degree vision.
There are security measures that should be incorporated into the system to make it more secure. But can we say with any degree of certainty that the system can ever be secure? No, it can never be secure, and that’s for sure.
If a man makes it, another man breaks it.
At the same time, we have to make sure that if there are intrusions or attacks, that we contain those attacks and we minimize their impacts.
NRGM: How do smart grids fit into the city of the future? Would a smart grid even be there, or would we transcend to something altogether new?
Farhangi: If I could let my imagination run wild, I would talk for the next two hours.
Realistically, what is going to be possible is that a city of a size like Groningen, or smaller cities in The Netherlands, will have a level of control over how energy is consumed. Ultimately, a smart city of the future should empower the users to be in control of their carbon footprint. In order to get there, a lot has to be done. From making the utilities work hand-in-hand with the consumers, and governments with both of them, to having full visibility over how energy is generated, stored and transported, to having other sectors integrated in this, including transportation, gas and district heating providers. But, given the fact that we live on a planet that has finite amount of resources and primary fuel, there is no choice but to go there. It is not a question of if, it is a question of when.
Recorded in Groningen, August 2014.
*A shortened version of this interview first appeared in a printed NRG Magazine, Nov. 2014. You can view it on ISSUU.