What is a microgrid?

Heather Smith • 7 November 2021

This resource is part of a series. See also, "What is a local battery?", "What is a Virtual Power Plant?" and "What do we need to think about as the grid decentralises?".

What is a microgrid?

It's a trendy term so it is prone to some overuse, often in slightly wrong situations.

There are a range of definitions. The Federal Government uses this definition,

An electricity supply arrangement that can (but may not always) function autonomously, and generates and supplies electricity to multiple customers.

I have decided to clarify some of the most important concepts about microgrids in this blog without dragging us too deep into a fight around different terms that could be used.

Centralised vs decentralised

Microgrids are a topic of discussion because we are moving toward an electricity system with generation sources, like wind and solar, scattered everywhere. The implications of decentralised energy is clearly something we need to think cleverly about. Some people have been singing the virtues of placing the supply of our electricity next to the demand for a long time so we need to make sure we capture those benefits. Microgrids can partly be defined by what they are not. They are not the main centralised electricity system. For many of use in Australia the centralised system serves most people in our state and is centred on our capital city. On the east coast, it is interconnected to the other states and runs as five markets sharing electricity between each other under the banner of the NEM or National Electricity Market.

In off-grid areas, smaller systems serve remote towns. These off-grid systems are microgrids. They exist because it is too expensive to run the poles and wires to connect these locations to the centralised grid. Western Australia runs at least 30 of them for most of its northern towns. Traditionally they have been designed with a diesel generator because they are too small for larger fossil-fuel power stations.

Renewable Energy microgrids are my main concern. The off-grid world was one of the first markets for solar energy because diesel is so expensive. Now that batteries are starting to become viable we can design off-grid microgrids with little or no reliance on the diesel generator.

Traditional microgrids are off-grid and (usually) not yet fully renewable

SCALE - microgrids are small to medium

Scale is therefore the first distinguishing feature of a microgrid. The International Renewable Energy Agency (IRENA) provides a clear definition of the different scales.

From IRENA: Off-Grid Renewable Energy Systems

For our purposes, in Australia:

  • anything that serves more than one home or business
  • in a region that is scaled from a street up to a suburb or a town

...can be called a microgrid. (because I refuse to call the larger ones minigrids)

Most Zone substations are sized from 5MW up to 30MW. They might service 3,000 homes at a time plus commercial businesses and some industrial users. A small town might be served by a single feeder and need 1-5MW of power. At a more local level your town centre or your street might need 100kW - 500kW of power. The size of your region will determine the generation that your microgrid will need.

Your microgrid is likely to be less than 5MW (compare with ~5,000MW for a centralised grid)

Disconnecting from the grid

Now that we know a remote town in outback Australia is the typical size and traditional use of a microgrid, let's talk about the scenario when your microgrid spends most of its time connected to the centralised grid. If we want to be specific we might distinguish these microgrids from off-grid microgrids by calling them grid-tied microgrids.

The key to "functioning autonomously" is disconnecting from the grid (especially when the grid is not supplying power). Connecting different electricity systems together is not a trivial exercise. When your solar system connects to the grid it is tuned to synchronise perfectly with the grid voltage and frequency so that there is not big surge of electricity across the connection point. Likewise, if the control of a microgrid is going to differ from the management of the main grid, it must be disconnected to do so. This is known as islanding. When the disconnection occurs, all the loads (and generators) downstream from the disconnection point (known as the Point of Common Coupling - POCC) become isolated from the main grid and can behave like an electrical 'island'.

Nowadays you can buy a battery system for your home that will keep your home powered up when the main grid is unavailable. You can even buy solar inverters that can do this without a battery! These are called stand alone power systems (SAPS) and they operate on the same principle as microgrids. The circuit breaker at the electricity meter feeding your whole home is opened to disconnect your home from the main grid, so that the solar or solar/battery system can balance its supply with your demand to operate autonomously. As the electricity user, this process can be seamless and you will not see the difference between operating in stand alone mode or main grid mode. You do need to make sure you don't demand more power than the system can supply.

Your disconnection point defines the region or 'island' that your microgrid needs to serve

Number of customers

While a microgrid operates on the same principle as a SAPS, it is more complex. There will be more than one customer to serve. There might be multiple generators and batteries. The biggest complexity might be that you don't own the disconnection point.

Embedded networks is the name given to systems, served from a single 'parent' meter through to multiple customers via privately owned electricity systems. This is often what happens at a caravan park, in a retirement village or in multistory offices or apartments. An embedded network operator would be running a microgrid if it could disconnect from the main grid at times and continue to supply electricity by managing internal generation and loads. Most embedded networks don't have microgrid capability and buy all the electricity from the main grid.

We are starting to see grid-tied microgrid projects that are operated by the electricity networks, for example Mallacoota. On the Heyfield project we will discover how difficult it might be to negotiate the ability to disconnect with the network. The network business is constrained by its regulatory obligations to supply electricity reliably and operating in island mode may also need to occur with specific performance expectations.

Microgrids are made more complex than a Stand Alone Power System by serving multiple customers and using a disconnection point owned by the network operator.

How to operate in island mode

Once you understand that a microgrid needs to supply adequate electricity at reliable voltage and frequency, you might be wondering how difficult that is. The key to a microgrid that relies primarily on solar and battery power is a grid-forming inverter.

Most solar systems have a grid-following inverter so they can't independently control how much electricity is pulled from the solar system. The grid-forming inverter knows exactly how much real power and reactive power the system needs and matches it perfectly. When a new load starts (ie you boil the kettle), the voltage will start to drop and the inverter will adjust the amount of electricity it draws from the solar (and battery) to ensure it meets the new demand.

This is easy to do when you have a surplus of electricity to supply. The investment required to always have a surplus will be high though. The key to operating a microgrid will be to ensure the loads don't demand more energy than is available, and some of this can be achieved if non-essential loads can be identified and turned off when the microgrid is in emergency-supply-only mode. The Mallacoota project is interesting because the battery system has already been delivered. The diesel generator remains for longer outages. The community is working toward extending the amount of time they can operate without resorting to diesel power.

First step - get a grid-forming inverter. Second step - work out minimum supply needs to optimise the size of your system.

What next?

I think one of the most interesting challenges facing us comes at the point that local resources start to displace network investment. Up until now, networks argue that they might not be needed sometimes, but they are needed most of the time so their investment levels remain the same. At some stage microgrid, battery or VPP investments will win the argument about supporting the grid so $$ originally earmarked for networks will head toward these investments instead. To me, this feels like a new social contract and I'm keen that we start talking about it. If you are interested, please read my next piece - "What do we need to think about as the grid decentralises?"



Also, please let me know in the comments if anything in this blog needs further clarity. And if you would like me to write more on this topic - thanks.