Could you provide a first-hand account of Superstorm Sandy, and how Princeton served the community through its microgrid?
The hurricane moved its way up the eastern seaboard of the United States, and as it approached campus, it took down trees, affecting power quality and reliability. Soon, it caused the utility power to go out, and as the voltage dropped, our plant shut down. However, the microgrid was able to self-restore, and we used the power to restart power supply to the campus. So technically, we were without power for only 15 minutes. The tricky situation here was getting administrative permission to isolate the campus from the rest of the power grid. Because we had cogeneration on site, we were able to separate and shut off the less important loads on campus. We ran separately from the power grid for days. We told people in the community, if their house is cold, we had set up cots and that they could come and rest at the University. However, most importantly, the first responders – that is, the firefighters and the police – were able to come to the university and get a meal and recharge their phones and radios. They were able to have meetings and decide what they should work on next, before getting back on to the field. The important takeaway here is that you don’t need every single place to have a microgrid, but if you have one occasionally, scattered throughout a community or region, it does make a huge difference.
Microgrids reportedly lower energy costs for customers. How much, though? Are the savings significant to overcome capex barriers?
The big savings come from the fact that microgrids enable cogeneration, thermal storage and other cost-saving opportunities. We generate a lot of power whenever the grid price is higher than our marginal cost to generate. We purchase a lot of power whenever the grid price is lower than our marginal cost to generate. That way, our campus customers enjoy the benefit of the lesser cost. Savings relate to dozens of factors and are very specific to the location, the grid and the energy needs being served. Payback may take from five to 15 years, depending on these factors. Or, under a power-purchase agreement, the savings may be immediate. In any case, though, the lifecycle savings are far more than the capex. Fewer microgrids would be built if they weren’t. Even utility customers outside the microgrid save a little money if they pay real-time utility rates, because our generation has the effect of slightly lowering the total net cost of power on the grid. That is, establishing a microgrid is a win-win for all customers, not a zero-sum game.
You describe microgrids as distributing risk into smaller pieces, whereby grid reliability is improved. Could you please elaborate?
Imagine there are two large utility generators serving a region. Each is capable of generating enough power to serve the grid by itself – that is, 100% redundancy. It’s easy to picture scenarios where those generators, or the substations or the wires between the plants and the ultimate customers are damaged. There are a few points with high vulnerability in the system that could interrupt service to large portions of the region.
Alternatively, if we reduce the size of the two plants and scatter several microgrids around the region, it is possible to have the same (or less) total installed generating capacity, while increasing reliability.
Microgrids, as per your description, reduce both energy use and peak demand and work well with CHP to greatly increase energy efficiency. This means they can widely be used as the system of choice, or do they work well only in some applications?
Microgrids would work everywhere, but they are not financially attractive and make no sense in some places. For instance, at home, on a tiny scale, you could buy two different-sized generators; however, you might spend five times as much if you run the air handler, the oven on self-cleaning mode, the welding machine and other appliances that use electricity. It’s not about what’s possible, but whether it is cost-effective and a sensible use of your financial resources. For instance, with regard to installation and maintenance, right now the utilities provide electricity to us, and you don’t need to worry about it. But if you decide to build yourself a microgrid, you need to begin doing what they (the utilities) have been doing. You need to make sure it’s safe and that you know how to operate and maintain it. You need to make sure that you have good fuel and that the fuel is of good quality. So by establishing a microgrid, you have taken on a lot more responsibilities and you get greater benefits. You have to ask yourself, “I am going to spend some money, but is that money going to be worth it?”
Could you elaborate on the economic motivation to conserve energy and how the scenario with regard to price and energy is different from the Middle East?
In Dubai, for instance, the price of electricity is the same all day and all night; whether you’re a residential customer or commercial, it’s the same price. In our state, in New Jersey, the price of power for residential customers never changes, but for commercial purposes it changes dramatically, as fast as five minutes. It’s not rigorously a demand charge but more of an energy charge that changes every five minutes. In the middle of the night I might pay two cents per kilowatt-hour, and in the day I may pay 25 cents per kilowatt-hour, so it could be 10 times as much. We have a very strong economic motivation and use as much energy as we can during the night in order to avoid the amount of energy purchased during the day.
Could you elaborate on the University’s control platform, which reportedly works relative to the energy needs of the campus in accordance with grid and weather conditions to forecast the corresponding load?
The control system advices us and helps us predict when electricity is going to be expensive and cheap. It is a combination of a software along with many meters, where we look at temperatures, pressures, flows and energy use through all the major equipment on campus. It is not rigorously an IoT-powered system, because when I think of IoT I understand it to be hundreds and thousands of data points connected to a system. In our system, we have a few hundred data points. I mean one temperature signal, a pressure signal, while the rest are sensors. The sensors are not scattered and are not radio transmitters; they are Bluetooth-enabled and hardwired directly back to the campus.
You say that existing generation assets can be operated in new ways for additional revenue with little capital investment. Could you please elaborate?
Yes. We built our cogeneration system in 1996. It wasn’t until 2003 that power was deregulated in our state of New Jersey. At that time, it became more lucrative to generate more power during the day and less power at night. We used the same asset fewer hours per year and generated more savings. More recently, the power grid has established a market for frequency regulation. By modulating the output of our existing gas turbine in response to a grid frequency signal, we are able to help support the local grid frequency. We get compensated for this activity at about 3x the price of power. It is a new revenue stream, same asset. It is a minor investment in controls.
Microgrids provide self-sufficiency and resilience especially in emergency situations. How can they be applied in the United Arab Emirates?
We put in cogeneration not because of reliability issues but in order to save us money. A collateral benefit is that it also gives us reliability and resilience in crises. I believe that the power supply in the United Arab Emirates is very reliable and very good, so there’s nothing to take away from that. But every once in a while, we still expect that there might be a problem, something that might take out the energy supply, and it would be nice to have spots of enhanced reliability. Having microgrids at critical locations, such as police centres, firefighting stations and hospitals, is crucial. The United Arab Emirates could benefit from cogeneration, not because the power is unreliable but firstly to save money and also to be more efficient. The power plant by itself might be 25-45% efficient, but if you do cogeneration, you could even do 75-80% efficiency. Hence, I am not concerned with who owns it; it could be a government or a utility or even a privately owned microgrid. Even with technology, you can use chillers, gas turbines, jet engines or diesel fuel.