After a week’s hiatus, I’m back with what I promise to be the final installment in the series. I know I’ve promised this in the past, but this will be the last one. (Don’t cry for me, Argentina) There are plenty of other stories about electricity to tell. So, this week we’re taking the installation of distributed generation and the deployment of smart meters as givens. We explore the impact that these developments will have on the grid, and what technologies will be necessary to make our electrical grid stronger and more reliable than ever.
Time for the standard disclaimer: These are forward looking statements, and as such, are highly speculative. Do with these thoughts what you will, but any losses or gains based on placing bets on my ideas are at your own risk. Though, if you want to send me part of your gains, I will gladly accept them. Let’s face it, if there was 100% certainty in the thoughts I’m about to share, I’d be placing my own bets, and probably wouldn’t be sharing these thoughts with you.)
Now to look into my crystal ball.
Problems with Distributed Generation
What I mistakenly referred to as “micro-grid” in the developed world in the previous article (How Electricity Grew Up, Part 5) was technically not a micro-grid. What I really went on to describe was distributed generation. Distributed generation does all of the wonderful things that I discussed last time, but distributed generation as it’s currently designed requires the electrical grid to be operating in about 99% of the cases. (There are some small grid-interactive battery-back-up applications out there, but they’re few and far between.)
So, why is that? The piece of equipment that transforms (or inverts) the electricity from a distributed generation source, either solar, wind, micro-hydro, or other, is called an inverter. One of the requirements of the National Electric Code for installing this equipment in a grid-interactive (connected to the electrical grid) fashion is that it is UL listed for that application. And, the UL specification for this application (UL 1741) requires that these inverters disconnect from the electric grid within milli-seconds of sensing that the electricity has been disconnected. This is done for safety reasons so that areas where the grid is having issue are not being back-fed by electricity. This protects both the electrical linemen that work to restore electricity in these areas, and anyone else who might be nearby. A break in any electrical wire can cause a ground fault, which usually comes with some sort of arc. This electrical arc is not only dangerous on its own, but can create very intense heat which can cause a fire. Shutting down the electrical grid in that area prevents this from happening while back-feeding the affected area would cause the problem to persist.
But, shutting down all the distributed generation in an area causes its own problems. Inverters are required to shut down when the voltage and frequency on the grid fluctuate beyond certain limits. This is how the inverters monitor the grid. But, in some instances, these fluctuations can be short lived. Let’s take a look at a low voltage event as representative of the multitude of events that can occur on the grid. A low voltage event is exactly that – an event where the voltage of the electric grid drops below the low level setting of the inverter. At that point, the inverter would shut off. But, let’s look at what causes a low voltage event. A low voltage event is caused by an instantaneous increase in demand on the electrical grid. This can be caused by a large piece of equipment or a large motor turning on at the end of a feeder circuit that needs that initial inrush of current to overcome the friction of the motor parts. The problem is that if the inverter shuts off while this is happening, the amount of electricity available to the motor decreases, which drops the voltage even further, thus compounding the problem.
Some larger inverters are have capabilities like low voltage ride through built into them already. This would allow the inverter to stay connected to the electrical grid so long as the low voltage event does not last any longer than a certain period of time.
The load from a large motor also causes the “reactive” power to lead or lag the “real” power produced by the grid. The comparison of “reactive” power to “real” power is known as the power factor, with a power factor of 1 being when “reactive” and “real” power are completely sync’d. This is a fairly advanced concept, so just remember that having “reactive” and “real” power too far out of sync can be a very bad thing. (Remember the black-out in the Mid-West in 2003 that crossed several states? A big culprit of that was not enough “reactive” power.) Many of the large inverters are also capable of power factor correction. Inverters have their power factors automatically set to 1. But, because of their technology, it’s just as easy for them to lead or lag the “reactive” power to the “real” power based on what is needed at the time.
Currently, reactive power is fed into the electric grid from the power plant all the way at the other end of the transmission and distribution grid. It’s like killing a fly with an artillery shell from one town away – it gets the job done, but it’s difficult to control precisely. By using the functionality already built into the inverter for power factor correction, an inverter much closer to the load causing the leading or lagging reactive power can compensate for that power factor issue in a much more precise and efficient manner.
Now that the inverters can use some of their capabilities to help stabilize the grid, let’s look at the next steps to making our electrical grid even more robust and redundant.
Micro-Grids (For Reals This Time Yo!)
The ultimate promise of distributed generation and smart grid (for me anyway) is the establishment of self-sustaining micro-grids as components. Imagine that each section of grid (think of it as each neighborhood) produces enough of its own electricity via distributed generation to handle all the critical loads. The number and amount of “critical loads” is going to change based on the neighborhood – think of a few lights and every refrigerator for a neighborhood of homes. How fantastic would that be? If a tree fell on the power lines between your neighborhood’s substation, and the power plant, you’d still be able to go about most of your life normally. (The only time this wouldn’t work is if the tree actually fell on the electrical lines in your neighborhood.)
So, why don’t we do this now? Well, for a couple reasons. First, there isn’t enough distributed generation installed to allow for the serving of critical loads. Second, as soon as the electrical grid goes down, the inverters would also shut off because they wouldn’t see the grid. Setting aside the first problem for a second, the answer to the second problem is to develop a sub-station or branch-feeder sized piece of equipment that can create the voltage and frequency necessary for the neighborhood micro-grid to work. This could potentially be installed at the substation level along with a transfer switch to disconnect that particular substation from the rest of the affected grid – thus making it into its own micro-grid. Once the rest of the grid is restored, the neighborhood micro-grid could connect back to the main grid.
This technology isn’t new. The same technology has been used for years for micro-grids in the developing world. It just needs to be made larger and more robust to handle a much large micro-grid. Considering they just installed three micro-grids capable of providing over 1 MW of power each, we’re not very far away at all. (Each home uses at most 48 kW (200A main x 240V = 48,000 W). So, 1 MW (1,000 kW) could fully power 21 homes. If you only power critical loads as suggested above, this number could be much higher.)
All of this is completely impossible without the additional data and monitoring that comes from the deployment of the smart grid. Understanding exactly where there are issues in the grid allows the utilities to dispatch the necessary functionalities from the inverters distributed across the grid and to isolate the different micro-grids in order to maintain power to the largest number of customers possible while isolating the problem.
Another technology that will be necessary to develop and deploy is going to be storage. Think of this as a battery, but there are several different types of storage technology currently available (fuel cells, compress air storage, fly wheels).
So, why is storage needed? The majority of distributed generation – solar, wind, and some micro-hydro – come from intermittent power sources. Intermittent power sources produce energy, but can’t be counted on to produce power. Think of water running through a hose. Energy is the amount of water that flows out of the hose over time. Power is the size of the hose – the most amount of water that can ever flow out of the hose at one time. In traditional electricity plants, you control the spigot. The energy to make electricity is held within the fuel (coal, natural gas, etc.), so the more you burn, the more electricity you produce. For intermittent sources of electricity, you don’t control the spigot, so you can’t turn up the spigot if you need more power. The trick is to store the water as it flows out of the hose, like in a bucket. That way, if you need more water at that particular moment, you can get make-up water from the bucket. That’s how it works, that’s why we need storage. And for the micro-grid vision I was discussing above, we need the storage contained within the micro-grid.
Value of T&D (Transmission & Distribution)
With all this talk of micro-grid, why do we even need the T&D infrastructure from our current grid? There is still a ton of value from the T&D infrastructure.
Firstly, the T&D infrastructure is absolutely critical for the transition from our current electrical grid to the future grid. We have ton of centralized power plants that are located far from where the electrical loads. We still to be able to send the power from those power plants to the loads.
Secondly, the T&D infrastructure provides the truly robust link between these various micro-grids. By maintaining that infrastructure, we can send power from one micro-grid to another. This adds redundancy, and makes each micro-grid more robust than if it were stand-alone.
Thirdly, as we transition to new and cleaner electricity power plants, we become more and dependent on building these power plants where those resources are best. With the current power plants, it’s fairly simple to move coal, natural gas, or nuclear fuel to the power plant. When you begin looking at solar, wind, or hydro plants, the power of those resources changes dramatically based on the geographic location. Large scale solar power plants are best in the southwest, wind is much better in very specific locations across the country and off the coast, and hydro works best where you have flowing water. The T&D infrastructure will allow for that electricity to be sent to where there is the demand.
It’s going to take a long time to transition from our current grid to the grid of tomorrow. But, we need to start thinking about it today. Over the next 20 years, several GW of coal plants will be shut down in the Mid-West. In the decades after that, several GW of nuclear plants will be shut down. All this power will need to be replaced. We don’t have the option to stay with the status quo. We can either replace the plants that are being shut down with new versions of what is being shuttered, maintaining the current problems and issues with our electrical grid. Or, we can establish a vision of the future, and use the money we’re going to be spending anyways to create something new that addresses all the problems we currently have with grid reliability, emissions, and plant siting.
Let’s take a deep breath, a step back from the problems, and dream about the future. It’s what Thomas Edison did many a time when he was around. So, let’s not just stand by his side, but let’s stand on his shoulders and envision, and build, something well beyond what he could imagine.