Posted on 17 July 2019

The Smart Grid Revolution

Free Bodo's Power Magazines!




GaN-based devices can overcome limitations

In December, at the climate conference in Copenhagen, the world again committed itself to producing a sizeable part of its energy from renewable sources. And to meet those quotas, the renewable electricity generation will have to grow sharply in the coming years.

By Jan Provoost, imec Science Editor and Johan Van Helleputte, imec Director Technology Office Energy


This, however, will put a lot of pressure on the electricity infrastructure, which is not yet ready to cope with the tidal wave of renewable energy that is in the making. It will require a radical rethinking, much innovation and new technology, and a major investment to get it up to par with the expected growth in renewables. It will require a revolution, leading to what is generally called the “Smart Grid”.

Imec, through its research on a.o. photovoltaics, lowpower sensors, and power electronics is helping the Smart Grid revolution to become reality. On the one hand, this research has shorterterm goals – think of further improving the energy-efficiency of photovoltaic systems, to reach grid parity even faster. But there are other areas were more fundamental breakthroughs are needed – for example exploring new materials and processes to for the next generation of power devices. These should be capable of coping with high voltages and high temperatures, and they should be cost-effective. The Smart Grid will rely on such innovative power devices for network stability and quality, load balancing and management, and for energy storage.

Energy: the telecom of tomorrow

Once we reach the targets set by the EC’s SET-plan, for example 20% of all energy generated by renewables in 2020, the landscape of energy generation will have changed completely. Next to the classic electricity plants, there will be millions of solar panels on rooftops, or private windmills and heat pumps that deliver electricity to the grid.

Compare this change to what has happened in telecommunications in the past 20 years. Like with energy now, there used to be only a few large telecom companies. They ran fairly homogeneous services, on networks and technology that changed only slowly. There was the phone, and you used it for voice communication, over a copper wire. Today, there are thousands of companies that offer telecom services through a mix of heterogeneous infrastructure and services, with very fast evolving underlying technologies. In this process, telecom has gone almost completely digital. This has blurred the boundaries with computing networks – the Internet – and has given both telecom and computing a sizeable boost. It has also fundamentally changed the telecom business models.

The same revolution is about to come to the energy market. With millions of heterogeneous energy generators – prosumers instead of consumers. And with a need for distributed and mobile storage points. With long- and short-distance delivery of energy, with real-time pricing, and a host of new services. And with a new grid that, while incorporating the old infrastructure, will function radically different: with much more intelligence and digital technology. With semiconductors that process electricity instead of information.

It will be a tremendous challenge to transform the electricity system worldwide into one of the most complex systems ever made by mankind. And to do this while assuring a continuity of services. This is not a green field technology deployment!

Image 1

The Internet of electricity: global, superdistributed, real-time balanced

The ideal place to produce renewable electricity is not always where it is consumed. Southern Europe, for example, or the vast sun-bathed spaces of Northern Africa are the ideal place to generate solar energy. But then you need a grid that can transport this energy to where it is needed, all the way up to the Polar circle, if needed. And this of course without losing too much energy on the way.

Another challenge is assure the quality of service, at any time and at any place, in a context where increasing amounts of electricity are generated intermittently and unpredictable. On a sunny summer’s day, more energy may be delivered to the net, at a moment when there may be less demand. Conversely, on a winter’s night, there may be a peak demand, but less supply. So the future grid will have to cope with peaks of supply and peaks of demand that are not necessary coinciding and that are much more volatile than today. This will require a sophisticated, fast and robust load balancing with new load management systems and storage approaches. A lot of intelligent transport, innovative storage and smart pricing will be needed to achieve this.

Third, there will no longer be a sharp distinction between producers and consumers of electricity. At any one time, a net consumer may become a small producer of electricity. And this can be implemented on the level of buildings, but even on the level of appliances – think of cars, for example. So all metering and balancing equipment should function both ways. In this context, utility providers will need, at any moment, fast and detailed information about the consumption and production profiles of all its customers.

For better balancing and peak-shaving, we will need real-time pricing – varying the price of electricity at any moment and place depending on demand and supply. Again, this calls for smart metering and fast feedback at the level of every appliance. Appliances, for example, that adapt their electricity use to the price of electricity at any moment, in some cases charging a battery when the price is low, and delivering energy from that same battery when the price is high.

In summary, electricity will be transported and balanced through a interlinked network of smart grids. And there will be potentially billions of consumers and producers. This calls for an infrastructure much like we have the Internet today for computing and telecommunication.

Power electronics at the heart – imec’s GaN-on-Si IIAP

The Smart Grid revolution will rely on billions of cheap smart meters, switches, sensors, actuators, convertors, invertors and batteries. Today, many of these devices are still too slow, too bulky, too expensive, and not fit for high-voltage or high-temperature use in tomorrow’s grid.

The high-voltage power devices that are already used are mostly based on Si MOSFET structures. However, for an increasing number of applications, their use is limited. GaN-based devices, for example, can overcome these limits. They show a unique combination of excellent transport properties and high electrical field operation capability. Not only from an operational point of view but also in terms of energy saving opportunities, such new generation of devices will become extremely important.

The issue, of course, is if such devices can be produced cost-effectively, on large wafers that is. Last year, imec in collaboration with AIXTRON has already shown that it is possible to grow a crack-free GaN layer (1.5µm thick) on 200mm wafers. This opens excellent perspectives for GaN-on-Si devices, especially from the perspective of cost versus performance.

To leverage its experience with GaN, and to allow partners to participate, imec has recently started a research program for GaN-on-Si device research. The goal is to develop high-voltage, low-loss, high-power switching devices based on large-diameter GaN-on-Si technology. Devices that, in the years to come, will be used as building blocks for the Smart Grid.



VN:F [1.9.17_1161]
Rating: 0.0/6 (0 votes cast)

This post was written by:

- who has written 791 posts on PowerGuru - Power Electronics Information Portal.

Contact the author

Leave a Response

You must be logged in to post a comment.