Energy Innovation in the Vault

Greetings from the Vault in Canberra, where an Energy Innovation Exhibition is being held. I was here a few weeks ago for an ACT Government Apartment Complex Electrification launch. This time there is about $20M in energy grants on offer from the ACT.

Some of the start-ups displayed seem readymade for the market. Examples are ways to repair old solar panels and an app which identifies where community batteries are needed. Others are a little more futuristic, such as a hydrogen powered speedboat (which looks like a USV from the Black Sea).

PS: * The technology does have defence application. At present there is concern over Australia's ability to patrol it's shores. One way is with small autonomous vessels. But these need more power than a battery can provide. This could be from hydrogen made on board, allowing the vessel to patrol for months.

Apartment Complex Electrification in a Vault With a Chef

Greetings from the "Vault" in Fishwick, Canberra, for the "Apartment Complex Electrification - Information Session". The vault is a windowless concrete structure, on a secured industrial site, looming like something from an action movie. Inside the Vault is an industrial kitchen, complete with a chef. As I walked in someone said "High Tom good to hear you on radio". 

This all distracted me from the public servants presenting on the Apartment Complex Electrification project. Some time ago I was on a working group advising the Chief Minister how to reduce energy use in the ACT. We focused on apartments as there were already programs for houses. But it turns out this is difficult for technical, legal and social reasons, thus the pilot program. 

After a few minutes it was explained the chef is demonstrating cooking on a low energy induction cooktop. The food was excellent.

The call for applications from apartment buildings to participate in the pilot will be released Monday.

E-waste not a major challenge with generativeAI

Wang, Chen, Zhang, Tzachor (2024) suggest that the demand for generative AI will create an e-waste problem of 1.2 to 5.0 million tons  for 2020 to 2030. They also suggest this could be made worse by political restriction on access to more advanced  efficient chips and rapid replacement of old hardware. I suggest the problem is not that large and energy use will remain a larger problem than e-waste.

One aspect the authors do not mention is the lack of price signals between server providers and the end user with current generative AI services. This is likely to be self correcting. Currently demand for generative AI is being generated by offering of free services to the public. As the user is not paying for the service and there is therefore no built in fee for responsible disposal of created e-waste, there may be a later problem. Some speculative AI ventures are likely to become bankrupt leaving a toxic legacy (similar to the mountains of scraped e-bikes left by failed startups). However, as users come to rely on Generative AI services, vendors will introduce charges, which can cover e-waste costs.

Currently generative AI server farms are using generic Graphic Processor Unit chips. These are the same chips used for cryptocurrency server farms. The environmental issues are similar with both. There have been articles about AI consuming as much power as small countries (just as there were for crypto). However, there is more of a mainstream use for AI, which will allow for better long term regulation of environmental effects. With its abundance of renewable energy sources and a stable regulatory environment, Australia could provide a popular location for AI centers. This would allow a small query to be sent across the world and answer sent back, effectively embedding the renewable energy in the answer. 

There may also be scope for reuse of older, slower, more energy using AI chips in locations with abundant renewable energy. As more efficient chips were installed close to the user in high energy cost countries, the old hips would be installed further away. Rather than store energy in batteries to run these chips, it may be cheaper to shut them down when the sun isn't shining & the wind isn't blowing. How to do this is something engineers and computer professionals can learn to optimize with specialist training (Worthington, 2012). 

Reference

Wang, P., Zhang, LY., Tzachor, A. et al. E-waste challenges of generative artificial intelligence. Nat Comput Sci (2024). https://doi.org/10.1038/s43588-024-00712-6

Worthington, T. (2012, July). A Green computing professional education course online: Designing and delivering a course in ICT sustainability using Internet and eBooks. In 2012 7th International Conference on Computer Science & Education (ICCSE) (pp. 263-266). IEEE. https://doi.org/10.1109/ICCSE.2012.6295070

Energy, Digital and People Skills for Australia

Australia 4.0 - Energy & Digital Skills Roundtable

Greetings from the Engineers Australia HQ, where I am taking part in the Australia 4.0 - Energy & Digital Skills Roundtable. This is organised by the Pearcey Foundation to work out what skills will be needed for the transition to zero emissions economy, and how to get them. This is recognising that even if you have the technology, you need people who can build and repair it. This is not just people bolting stuff together, but also those programming the systems to run it, thus my interest*.

After a welcome from the acting EA President, we had some statistics of the supply of engineers. Highlights were that Australia is dependent on importing trained engineers and it will take 70 years at the current rate to achieve parity between male and females. This is much the same as the computing profession. There is an obvious way to increase the number of domestically trained engineers: attract more females, but that will require changes to training and jobs. That could also solve another problem with technical training: the need to teach people skills. At the ANU I help teach computer students to work in teams in the Techlauncher program.

It took 30 minutes before AI got a mention in the presentations. This was a relief as for the last few years AI seemed to be in the title of every talk. ;-) The approach taken here was AI enhancing work.

Some of the good news was that engineering skills are transferable. This suggests that engineers could move to renewable industry from others, such as mining. The bad news is the reverse is also the case, with the mining industry in particular having a demand for staff and money to attract them. Also there are about 40% of qualified engineers working outside an engineering role who might be attracted back. Engineers Australia produced a Clean energy workforce capacity study submission (May 2023).

The roundtable switched to comedy with discussion of smart toasters talking to smart fridges. This had a serious side as a way to manage energy use and also as a potential privacy risk. 

A topic I want to raise with the roundtable is the government's recent restrictions on international students: will this reduce the supply of engineers and computer professionals? What can we do to increase the domestic supply? Microcredentials were mentioned and I wonder if they will help people already qualified in engineering or computing to transition to renewable energy. Shortly after I typed the last few sentences the chair invited me to speak, so I said it.

The next speaker proposed an updated energy risk assessment. An example given was the vulnerability Australia has importing the components for building renewable energy systems. Also it was claimed that in a few years there will be no one in government qualified to conduct such a study.

* Way back in 2008, the Australian Computer Society commissioned me to write a course on Green Computing. The ACS and ANU first ran the course in 2009 and it is still offered by Athabasca University in Canada, 15 years later. The course is still on the books at ANU, perhaps it is time to revive it.

Batteries Included for Dunkelflaute: Future of Neighborhood Batteries in Australia Conference

Greetings from the 2nd Future of Neighborhood Batteries in Australia conference, at the Australian National University in Canberra. The event started with what we didn't know, which is if neighborhood batteries would "scale", that is with lots of them. One question I suggest needs to be answered is if there is value in community involvement in neighborhood batteries, or should these be something they just use.

ps: The word of the day is "Dunkelflaute". 

ANU Energy Update 2023

Greetings from the ANU Energy Update 2023, from the Australian National University in Canberra. The mood of the event is relatively upbeat, with Australia progressing well with the transition of its electricity network to renewable energy, and in a strong position internationally to sell important materials, renewable energy, and green refined metals. Australia has also benefited from subsidies on solar panels and other renewable equipment in other countries. It is not all good news with Australian politics having ruled out a simple carbon tax.

Australia an Exporter of Renewable Energy Embedded in AI Models?

The energy use of AI was discussed by Schwartz, Dodge, Smith, and Etzioni (2020). The authors noted that Amazon AWS was 50% powered by renewable energy. Perhaps Australia could become an exporter of renewable energy embedded in Green AI models.

In 2017 my colleagues at ANU initiated a project to cover large areas of northern Australia with solar panels, and export the energy. The university invested $10m in research, and some of it is now being commercialized.

Options investigated included a cable to Singapore, & synthetic fuel in tankers. But an alternative was "Green Steel". The energy would be used to refine iron ore in Australia. The embedded energy would, in effect, be exported in the steel. The steel is much easier to ship than electrons, hydrogen, or ammonia. The same thing might be done with AI.

The world would send AI requests to Australia. These would be forwarded
to data-centers located at high capacity grid connections to solar and wind farms. AI models would be trained using the renewable energy, making an "AI battery", with the embedded energy stored in the models.

Include a Global Power Grid in the Australian Energy Transition Research Plan

Greetings from the Australian Council of Learned Academies (ACOLA) Australian Energy Transition Research Plan webinar. Professor Ken Baldwin of the ANU,  used the often repeated phrase that solar power is only available "when the sun shines". It occurred to me that perhaps we should be bolder in our ambitions. The sun actually shines all the time, it is just the rotation of the earth shades part of the planet, plus local clouds. A cable around the world could distribute power from where the sun is shining to were it is not. This would be similar to Arthur C. Clarke's promotion of geostationary satellites for communications. Before this it was assumed satellites would have limited use for communications,  as they would only be over a point on the earth for a short time. It took considerable time, and investment, for the geostationary satellite to be a reality, but the concept helped legitimize satellite communications. A bold concept of a world-circling power grid might have a similar effect.

Professor Sue Richardson expressed concern about the capacity to train enough people for the renewable energy industry. However, we have the capacity to do such training online, which has been boosted by COVID-19 forcing both vocational and university education online. One of my ANU students was in a hurry and completed my 12 week green computing course in 4 weeks. ;-)

Will an Electric Ute Plugged into a Shed Save the Australian Electricity Grid?

Car charger at ANU

On my way to the cafe opposite my office at the Australian National University this morning, I stumbled across a new shed, holding the future of Australian motoring and perhaps the salvation of our electricity grid. This shed holds an electric car charger, as part of a vehicle-to-load (V2G) trial by the Distributed Energy Resource Laboratory (DER-Lab). Not only can an electric car be charged, but the energy fed back to the grid later.

I like the very understated, practical and Australian style of the ANU charging station. It is a plain grey steel shed, with a corrugated steel roof, like a rural bus shelter. So this approach could be called V2S: Vehicle to Shed. Add solar panels to the roof for Silicon Shed. ;-) The corrugated steel roofed shed was made a design icon by architect Glenn Murcutt, with his Lerida Estate Winery building outside Canberra.

Using a car to power the grid might seem a waste of money: why not just buy a stationary battery? However, range anxiety, along with Australians love of large vehicles, plus business finance and tax deductions, may come to the rescue of the grid.

The average commute is less than 40 km per day, but consumers are demanding electric cars with ranges of more than 200 km. Drivers worry that they will run out of energy and be stranded (so called range anxiety). So the electric cars will have excess battery capacity which could be used to store energy for a home or the grid.

Some newer vehicles come equipped with a plug to not only charge the car, but also return surplus energy back to the household (V2H), or the grid (V2G). A medium sized electric car has sufficient battery capacity to power the average Australian home for a day and still have more than enough power for the daily of commute. This could be particularly useful in Canberra, which has net renewable energy, but limited storage.

The question then is, having purchased an electric car with far more battery capacity than needed day to day, can you convince the consumer to make the surplus capacity available for the grid? One of my students is researching how to Cut City Air Pollution Using ICT.

The use of V2G is an example of where technology works, but how do you get people to use it? What type of interface will make V2G practical? Does this just need a button on the charger to request a top-up for extended driving, or does there have to be an app, where this can be scheduled? Could the system check the family's online schedule and predict when the car will be needed? Does the car owner need a cash incentive to plug in "Plug in now and receive a $5 cash bonus".

A problem with V2G is ensuring vehicles are plugged into be charged. Peak solar power is produced in the middle of the day, when a commuter's vehicle will not be at home. However, I suggest the vehicle could be plugged in at home, charging overnight from wind farms. After providing power for breakfast (the morning peak demand), the vehicle would be unplugged for the drive to work. Arriving before the mid-day solar peak, the vehicle would spend several hours charging. As the sun sets, the vehicle would be unplugged to be driven home, where it would power the evening peak demand. 

This scenario would require two charging stations for each vehicle: one at home and one at work. It has been assumed that domestic chargers could be low current, as they can charge overnight, whereas public charging stations require high current for driver convenience. An alternative would be to have a higher capacity V2G station at home and a lower capacity, lower cost, charge-only station at work (something as simple as an ordinary power point). There would be a cost in installing chargers, and a problem if drivers wanted to use their car at lunchtime, which is the solar peak.

There has been some concern that electric cars would not suit Australia conditions, where large vehicles with long range are needed, as exemplified by the trade's Ute.  However, the world's largest, most powerful land vehicles are electrically powered and several companies are preparing large SUVs and Utes for sale. The large batteries in these vehicles could power a home for several days and offer a solution for grid storage.

Australia folklore has it that the Ute was developed for farmers to take produce to market during the week, and the spouse to church on the weekend. Dual cab utes are now very popular to take tools to work during the week and carry the family's sporting equipment on the weekend. What is less well know are the financial reasons for the ute: classified as a business vehicle, the farmer could get a business loan for it. So will today's tradie buy an electric ute with V2G, using a business loan, and tax deductions? This could be called U2G: Ute to Grid, or U2S: Ute to Shed. ;-)

Before dismissing the idea of a ute powered grid, consider that Australia leads the world in solar panels on domestic rooftops. While other countries had policies for large solar farms, Australia almost by accident, has one quarter of homes with their own solar panels. Having shown a willingness to invest in home energy, will the same householders embrace V2G, to the extent needed to support the grid?

If half of Australia's homes with PV panels were each plugged in to a vehicle providing 20 kWh of storage for the grid, that would be 20 GWh in total. This is one hundred times the capacity of the SA Big Battery (officially known as the Hornsdale Power Reserve). 

However, V2G also presents a challenge to the current energy providers. With enough storage for several days use, available essentially for free, a household could decide to provide its own energy from their rooftop solar for all but a few days a year. Energy companies will receive no revenue from selling energy for most of the year, as it is used is "behind the meter", but then have to provide on demand for a few days. Companies will need to provide households an incentive to have their batteries for part of a grid system.