The device sitting atop the utility pole outside your house could begin to look more like a switching power supply than a transformer in years to come. The result would be big changes in how residential solar panels and electric vehicle chargers hook into the utility grid.
The distribution transformers that feed power from utility lines to homes and businesses are soon going to change. Their replacements are solid-state devices that could double as inverters for solar panels and as chargers for electric vehicles, eliminating the need for separate devices providing these functions.
Today’s distribution transformers do little more than convert kilovolt-level power to a low voltage, useful for powering buildings. The solid-state transformers (SSTs) that replace them are likely to be adept at sending power back to utilities from residential solar panels and other sources of distributed energy.
SSTs won’t just route energy back and forth. Once decentralized solar PV generation becomes commonplace, the highly erratic nature of PV power could compromise grid stability. SSTs will play a role in regulating grid voltage and stabilizing operations in this scenario. Protoype SSTs on the drawing board also have a simple DC interface to integrate energy storage devices that will be useful for smoothing the PV output.
To manage all these tasks, tomorrow’s power distribution devices will effectively serve as local power grid managers, able to adjust the power factor of the buildings they serve, as well as managing the loads and distributed power sources in ways that prevent difficulties for utility substations.
For a glimpse of what tomorrow’s SSTs could look like, consider the work going on at the FREEDM (Future Renewable Electric Energy Delivery and Management) Systems Center associated with NC State University. Working under a grant from the National Science Foundation, the center has developed electronics aimed at handling bidirectional power flows as would arise in communities with a high concentration of renewable energy sources.
“The core of the research has been to develop an SST as a key enabling technology for the utility architecture of the future,” said Alex Huang, NC State distinguished professor and the center’s founder.
Under development at NC State’s FREEDM Center is the A 7.2 kV/25 kVA SST prototype modeled here.
The work at FREEDM depends heavily on its research to develop SiC semiconductors able to operate at high voltages. Present-day SiC devices can handle up to 1.2-kV applications and Huang expects NC State researchers will be able to devise SiC circuits handling up to 30 kV.
Huang’s group is approaching SST designs as special-cases of switch-mode power supply topologies. In that regard, they are devising pulse-width-modulation control schemes operating at frequencies in the range of 30 to 50 kHz.
The attraction of SiC semiconductors is that they make possible switching power supply topologies that are simpler and more efficient than those built with conventional silicon power devices. Still, there are a lot of technical challenges to fielding a pure SST able to replace the devices now sitting atop utility poles outside most houses. “I think it will take some time before a medium-voltage version of an SST will come to market,” said Huang.
Some of the challenges of commercializing SiC SSTs are at the circuit level. Because medium-voltage SiC power circuits are relatively new, there are still lessons to be learned about handling sharp current and voltage spikes and how to mitigate the resulting electromagnetic interference. The high isolation voltages involved can also be challenging. “Here we have some good solutions. We can take isolation voltage up to 20 or 30 kV, but this also drives up costs,” Huang said.
One lesson to come out of SST development efforts is that the industry may have to rethink expectations about the lifespan of such equipment. “Technology is changing so fast that a 10-year lifespan might be enough rather than the 30-year life of today’s transformers,” said Huang.
And though prototype SSTs have been described at technical conferences, their design details are far from being set in stone. “Many of the SST circuits we’ve demonstrated use well-understood circuit topologies,” said Huang. “For SiC semiconductors, those topologies might have to change significantly to make the best use of these devices.”
Though SSTs are still a research topic, the first steps toward their commercialization are already in place. Electric utility Duke Energy, for example, is now testing advanced electrical distribution devices that approximate many of the features of solid-state distribution transformers. Some of the devices in the Duke pilot project are made by a firm called GridBridge, a 2012 spin-out from the NC State labs. Huang, who is also a GridBridge cofounder, said the firm’s version of an SST isn’t as technologically aggressive as the SiC prototypes coming out of the FREEDM Lab. GridBridge’s devices, dubbed grid energy routers, capture many of the functions of an SST without replacing the transformer itself with solid-state electronics.
In field tests with Duke, GridBridge grid energy routers will demonstrate grid management techniques that include conservation voltage reduction (basically managing voltage back to the substation to maximize efficiency) and volt/VAR optimization (basically managing line voltage and inserting either capacitive or inductive reactive power to optimize efficiency). These adjustments can take place either to meet conditions programmed into each grid energy router or in response to commands coming from the utility. The GridBridge equipment will also help manage power line quality by canceling out harmonics and damping transients.
“We want to create reliability so buildings or residences getting power from a substation aren’t affected by events upstream,” said GridBridge CEO Chad Eckhardt. He also says that GridBridge has a product roadmap for how its grid energy routers will evolve as technology advances.
In the SST test environment at N.C. State’s FREEDM Center, the prototype SST sits on a table in the background. The big step-up transformer next to it serves as a way of generating grid-level voltages for testing the SST.
One thing is for sure: There’s a big market for the kind of smart transformers companies like GridBridge are making. “We’ve seen a strong response from utilities,” said Eckhardt. “An average utility might buy 20,000 distribution transformers annually and we think some of these will start to be energy router-type devices in the relatively near future.”
But much remains to be done before energy routers can be deployed widely. Equipment standards, for example, will need to evolve so they can be applied to solid-state transformers hooked up to the grid. Today, one of the principle standards for distribution transformers is IEEE C57.12.20-2011, but it is meant to cover only mineral-oil-immersed transformers, not the type of solid-state hybrids created by companies such as GridBridge.
Moreover, standards organizations are only now beginning to think about modernizing some of the applicable requirements for grid-connected gear. In the case of solid-state distribution transformers, Matt Wilkowski, chair of the IEEE Power Electronics Society’s standards committee, has been dormant. “We are trying to reactivate the committee now. The initial steps will be to create a forum of experts from utilities and energy router technologists who will determine what standards should apply and what specifics are needed.”
Wilkowski said there are many similarities between grid energy routers and electric vehicle chargers, at least in terms of requirements. So these similarities might serve as a starting point for new standards. SPW