Who or what will win the battle for the grid edge?



There’s a battle on for control of my solar PV system, and the electrons I generate on my rooftop, from the panels I paid for so that my family can save money and be a bit greener too.

Depending on who wins, the outcome will be either massively expensive technological overkill to allow the ‘system’ to take control, or lower-impact but still super-smart solutions that will empower the ‘consumer’.

Here’s my take on it. If you are anything like me, you’ll hate how the energy industry comes up with insider jargon that totally excludes people from the electricity system. A hot example at the moment is ‘the grid edge’.

By that, they mean my home, your home and millions of other electricity consumers, and all of our machines, appliances and devices that use electricity supplied through the utility billing meter, and sometimes equipment that we own that generates or stores electricity within our homes as well.

So in industry-speak ‘the grid edge’ is ‘behind the meter’, and it has become of increasing importance in recent years because of another bit of the lingo, ‘distributed energy resources’, or DER.

To give you an idea of the fully width of the DER definition, the Australian Renewable Energy Agency (ARENA) defines it thus:

‘DER encompasses behind-the-meter technologies such as rooftop solar, home batteries, inverters, controllable loads both in homes and commercial and industrial facilities, electric vehicle charging points, smart appliances and systems (such as fridges, air conditioning systems, hot water heaters and pool pumps) as well as relevant enablers such as smart meters and data services.’

The big challenge for the electricity system (aka the grid), which traditionally has been largely blind to what’s happening at its edge, is that it will now be impacted on by what people are doing at the extremities. It needs to protect its core assets, its profits and its reputation, plus keep the lights on.

Currently there is really only one main thesis for a new ‘edge-responsive’ grid. This involves embracing the proliferation of mainly residential DER, and then getting some degree of system-level control over these new, highly-decentralised sources of electrons. In Australia, and around the world, many hundreds of millions if not billions of dollars are being spent on this, in a bid to tame this new electrical frontier at the edge, now that large-scale generation like coal-fired power stations is declining at the centre.

So how to do this? Well the prevailing wisdom has produced a further extension of the jargon, with the addition of the phrase ‘grid edge computing’. In layman’s terms, it means that extra computing processing power needs to be installed at the home so that the solar inverter, and often a battery inverter too, can be monitored and controlled primarily on site, while still being linked to a centralised control source through the cloud for coordination with the grid, or at least parts of it.

Get this right, with many homes organised as a fleet, and you ascend the lexicon to ‘virtual power plant’, or VPP.  Effectively it means that the individual assets like solar panels, batteries and appliances that I/we/you as a consumer buy and install in our houses will be at the orchestrated remote beck and call of the old energy industry, as it struggles to help balance the grid in the 21st century, and to preserve relevance as reliable purveyors of this essential service.

In theory, I admit, this is not a ridiculous concept. However, in practice, really?  I have a couple of major in-principle issues with a grid edge computing focus.

Firstly, let’s look at the industry that’s trying to do this. Trying to leap straight to grid edge computing and virtual power plants is like trying to do a manned Mars launch before ever getting a rocket out of the earth’s atmosphere.


In electricity terms, energy efficiency is equivalent to getting out of the atmosphere. But hardly anyone is talking about it, instead being captivated by the battery and VPP hype. Yet efficiency remains a big, value-adding low-hanging fruit opportunity. At best estimates, there is up to 30%+ efficiency gain to be had at residential level, and at worst 9-10%. This is outright energy waste or unproductive usage. The best saved kWh is the kWh never used in the first place. So before we try the Mars shot, embracing the entire grid edge computing agenda, let’s spend a little bit of effort getting our rocket licence by removing the 10-30% energy waste first.

Secondly, residential premises are complex and unknown beasts, primarily because humans live in them and humans have a bad habit of doing whatever they want in their own homes.  Households in Australia also have prior form in the adoption of new consumer-facing tech, and will routinely adopt the latest and greatest shiny tech object (think offerings from the likes of Apple, Amazon and Google), and therefore the grid edge is going to be changing at a faster rate than our traditional energy players can handle.

Thirdly, where are the tech giants?  Nowhere to be seen (yet), however they will be.  Once data is freed up from the grid edge residences, the algorithms developed by the likes of Google, Apple and Amazon will easily ingest this data and create a series  of do-it-for solutions – designed on inputs, and streaming outputs. What does this mean? The tech giants, as many of you have experienced, have well mastered complex algorithms around the inputs we provide them, whether it is Siri, Google Home or Amazon Alexa, these engines learn from everything we do and provide solutions?  Why should the energy grid not fall into this tech?

The energy industry, as I have stated in previous blog posts, always says it is a special case, based on its essential service status, and the political imperative to avert blackouts. Is it special, however?  Or is it merely as essential as our other essential services, like transport, internet and telephone? These are all industries that have been disrupted and interrupted by large tech plays, and ultimately energy will be no different.

The engineers within energy will say that the grid edge must be operated within milliseconds and must have direct control at household level, hence the desire to throw more energy-specific computing power into homes. No surprise here, that engineers want engineered solutions.

Yet this seems like a big jump from where we are today, which is a heavily industry-centric model of energy deployment, to a utopian (for the engineers) future of extended control into the home. The energy technologists will say that what you need for the grid edge is guaranteed communications to the grid, communications into the home (WiFi) real-time energy signatures, and an always powered-on device.  With this you can effectively monitor the heartbeat of any residence. With the data from this, you know the state of play off the home at a ground level.

I suspect this is technology overkill, which is only being seriously contemplated because the old industry is increasingly desperate.

An alternative approach will utilise low-cost IoT sensors and controllers, with cloud-level computing driving alerts and actions. For example, we can make a simple technology policy change to insist that all major appliances such as air-conditioners, heaters and hot water systems include WiFi and an industry standard API. This will  allow an algorithmic cloud-based decision process.

It doesn’t matter if one house can respond in milliseconds, while another house takes minutes or hours. As long as the status, response time and appliance landscape are known, algorithms can predict and inform households on what actions need to be taken to save money, most importantly, and secondly to react to financial incentives  from the grid to reduce demand.


Fourthly, networks want control.  This is, again, a matter of trying to replicate the old into the new. The concept of controlling every major appliance within a house is laughable.  Changeover of these appliances is in the decades, not days or weeks, and easy interoperability is a pipe dream. So while getting appliances to a) be smart, and b) work smartly together, remains illusively futuristic, there is a low-tech option – interacting with humans and their fingers to control their sites, and using real-time data and communications to coordinate many sites, using appropriate incentives.

Technology-wise, it is still early days in some ways for batteries and even solar inverters, however, the brand landscape is hoping to continue to expand. It is on track to become more and more complex to interact with, not less, thus making any notion of a nationalised control framework unsustainable.

In the middle of all this, alongside the missing-in-action efficiency piece, Australia now has over two million mainly residential rooftop solar systems that mostly are unmonitored, and invisible to network and grid operators.  We have no idea of their state of performance, operation, or whether they are even still switched on. Yet by knowing the state of things, we can vastly improve outcomes without jumping to the control.

If energy efficiency gets our rocket out of the atmosphere, then using data to mobilise our national solar fleet is like an achievable moon mission. Mars can wait.


GAVIN DIETZ is CEO of Wattwatchers Digital Energy and its parent company, the Energy Saving Networks Group. He is a mad-keen home technologist with a passion for empowering consumers through their energy data.