Supply voltages, transmission losses, long-term reliability, IEEE standards, circuit simulations – those are familiar topics for those working in the semiconductor industry. But they are of equally high interest to another group: those working to improve the electricity grid.
Certainly everyone has heard by now that "The Grid," at least in the U.S., is fairly antiquated. DOE Secretary Stephen Chu recently quipped that if Thomas Edison were to suddenly come back to life he’d have no idea how an iphone works, but he’d be quite familiar with all the elements of the electricity grid. They’ve hardly changed since Edison designed the first grid for New York city more than 100 years ago.
Certainly many people – perhaps too many – have left the semiconductor industry to pursue a career in the exciting world of photovoltaics, where the substrate is usually silicon and the processes of doping, annealing, metallization and packaging are familiar. The output of the device, however, is not processed signals, but raw power.
The world has embraced photovoltaics and other renewable energy sources such as wind, hydro, biomass and geothermal. One of the concerns moving forward is that the best of these sources – PV and wind – are inherently intermittent. Proponents of PV like to point out that although PV is intermittent (due to clouds and of course darkness), it’s actually highly predictable. Clouds don’t cause that much variability if the PV is spread out over a wide enough area, and because they are visible, it’s relatively straightforward to predict the impact on power generation on a short-term basis and even easier to predict the amount of power that will be generated the next day based on weather reports. That’s fine because power markets operate on a day to day basis.
One way to balance out that intermittency is through energy storage. Analysts see a strong, upcoming demand for energy storage as part of the grid. This will likely be a combination of some kind of central storage (i.e., a 20MW flywheel installation near a power generation station) and distributed storage (i.e., batteries next to the familiar green transformers in people’s yards). These types of energy storage are primarily driven by a need on the part of utilities for load balancing, since it’s expensive for them to constantly adjust the output of traditional power generation systems as the load varies. Energy storage may even allow them to offset or delay the requirement of additional power plants, such as a gas-fired "peaker" plants which are notoriously expensive.
In some markets, there is also value for companies and people on "the other side of the meter" to buy and store power when it is least expensive, and use the stored power during peak demand when prices are highest. Electric vehicles will also come into play, in part by helping to advance battery technology, but also by actually becoming part of the smart grid. Andy Chu, director of marketing at A123 Systems (Watertown, MA) envisions a time when utilities are so linked into the grid that they can monitor and control electric vehicle battery chargers during the night, and charge them quickly or slowly so as to optimize the load/generation equation.
This vision of the smart grid with renewable sources and energy storage working in harmony is complicated. The U.S. electric industry includes over 3,100 electric utilities. Investor owned utilities are privately-owned, represent 8% of the total, approximately 75% of generation capability and revenue. There are 2,009 municipal utilities, supplying approximately 10% of the generating capability and 15% of retail revenue. There are 912 cooperatives, operating in 47 States, accounting for 9% of total revenue and around 4% of generation.
What needs to happen to get them to work together is simple: standardization. This is where I believe those in the semiconductor industry can make an impact — volunteering to participate in standards committees. One standard of importance is IEEE P1547.8, which is focused on high-penetration, grid-connected photovoltaic technology. Among the issues being discussed: active voltage regulation, voltage and frequency ride-through, frequency trip settings (under/over voltage), operation under fault conditions, switching, power quality, monitoring and control, and dynamically controlled inverters. Getting involved is easy: check out http://www.nrel.gov/eis/high_penetration_pv_wkshp_2010.html for more information. Your experience is needed!