Anyone who is familiar with the early history of the utility industry in the U.S. knows that a battle raged over whether the electric grid should be Direct Current (DC) or Alternating Current (AC) (http://energy.gov/articles/war-currents-ac-vs-dc-power). Thomas Edison and General Electric favored DC while Nikola Tesla and Westinghouse thought AC would be best. Ultimately, AC power won out because of the ease of changing voltage levels with a transformer and use of polyphase induction motors.
Today, with few exceptions, the electric grid is predominately AC. However, it appears that DC power grids may be on the verge of a comeback of sorts. Digital equipment, solar PV, storage batteries, electric vehicles and other end-use devices all require DC power. Data centers are chockfull of such devices and for several years there has been a movement toward DC data centers (http://www.greentechmedia.com/articles/read/a-hidden-benefit-of-dc-power-real-estate). Additionally, others are calling for our houses to embrace DC power, as well (http://www.greentechmedia.com/articles/read/Time-to-Rethink-the-Use-of-DC-Power-for-the-Energy-Smart-Home). Some say there is a larger role for DC power on our grids to improve stability and enhance reliability.
With the growing deployment of DC power and the ongoing advances in power electronics, the question is: What will be the role of DC power in the Utility of the Future?
Tesla, Westinghouse and Edison were three giants in the industry who fought a battle for which standards to use. An elephant died in the pursuit of a single standard for the grid (killed in an attempt to show that AC power was inherently dangerous). Eventually, it was an electric motor that made the difference.
Today the difference may be simple silicon-based devices. No elephant will likely die in the process, but the outcome may be very different than it was over 100 years ago. In the scenario where a home has photovoltaic (PV) on the roof and an electric vehicle (EV) in the garage, the overall efficiency of the system is much higher if the electricity from the roof never is converted to AC. Put a battery in the middle and a pure DC system between the roof, battery and the EV is much more efficient than going DC-AC-DC-AC-DC. Just the conversion losses alone sap the energy out of the system.
The question is not whether DC will have a role in the system moving forward, the question is where and how fast. EMerge, an industry alliance that is focused on DC systems in buildings, has completed DC standards and has certified more than 400 products that are DC related to go in commercial and industrial buildings. In the case of commercial buildings, the use of 24-V DC for lighting makes business sense because any maintenance worker can run 24-V lighting systems, removing the need for an electrician to be called to do the wiring in an office move. Add that the LEDs used in the lighting systems are much more efficient than the fluorescents typically used in offices and doing conversion of AC to DC makes business sense. Other devices may not be as useful in the office when the power is delivered as AC to the building, but as Walmart and others begin to put large PV arrays on most buildings, the use of DC makes more business sense.
DC data centers are a reality where there is little or no AC in the whole facility. AC to DC conversion from the grid is done outside the building to limit the amount of heat released from the conversion in the building. Some are on the drawing board that will exceed 1 GW of DC power in the data center, most are much more modest facilities. The backup generation and batteries that provide the reliability for the data centers are wired directly into the DC infrastructure of the building.
In the case of homes, it is surprising how many devices use DC as their primary form of electricity, from microwaves to some washers/dryers and refrigerators, DC is becoming the choice for many appliance manufacturers. In the IEEE’s [email protected] activity, more than 1,000 different appliances that are popular have been characterized by the research team. [email protected] has membership from over 400 companies, universities and government agencies covering more than 50 countries. The team is working on draft phase I reports that are due by the end of 2015. In the meantime, EMerge has started a standards effort for DC devices in homes. The National Electric Code needs no changes to support low-voltage DC wiring in the home, to the point that Next Energy in Detroit has a DC home that is up and running as a test lab. The home has PV on the roof, batteries, EV chargers, LED lighting, an all DC kitchen and entertainment system. In fact, almost anything you could want in a home is available in DC.
Obviously, when one looks at the transmission system DC is alive and well in the links that bring hydro-electric power from Québec to the northeastern U.S. and in the grid in ERCOT, as well as new transmission design in countries like China.
In discussions of microgrids, at least in the IEEE paper submissions, DC as a power choice is winning over AC as a power choice.
The IEEE Power Electronics Society (PELS) has a draft specification for an energy router, a device that can on the output side send out either AC or DC power (or both) and on the input side take AC power. Companies like GridCo Systems already have energy routers in the field being tested by investor-owned utilities, not only to deliver DC to customers but to assist with voltage management, power quality and other localized issues on the AC front. Rumor has it that at least one very large equipment manufacturer is testing DC-related distribution equipment in the field. Utility-scale renewables, in many cases, have a DC output and then are converted to AC after the power is generated.
The question in the distribution grid will be not one of AC or DC, but what the right mix of both is over the next 20 to 40 years. It may start slow and close with a solar farm delivering DC power to a building across the street to run fast chargers, or a solar farm being built at a rest stop to provide charging for EVs. From there, it may turn into a college campus or a commercial development getting some or all of their power delivered from a combination of rooftop and terrestrial solar via DC. Eventually, there may be a mix of AC and DC delivered into neighborhoods. What we don’t know is will that power be delivered on new infrastructure, on existing infrastructure, via battery on trailers or what.
The other question is given the issues with switching, routing and breaking DC, will we attempt to mix AC and DC on the same wires, on parallel wires, or in complete separate infrastructure owned potentially by different owners?
There is a lot of movement on DC today from the lowest to the highest voltages. There are more questions than answers from a distribution utility point of view. At the highest voltages, we have solutions the same can be said for the lowest (3, 5, 12 and 24 volts) in between is the space that will take work to figure out.
The industry is changing. What role are you going to play?
The macro-grid (actually three grids) in the U.S. is predominantly HVAC and will continue as such. But HVDC will make inroads for long distances as fuels for electricity transform over the next several years, and for undersea and underground advantages (e.g., aesthetics, cable technology, and no short-circuit contribution headaches for already very strong urban networks!). Having said that, the grid in aggregate will remain with an HVAC dominance because of the legacy systems and high degree of interconnection (mesh) already in place.
Of course, the three grids of the U.S. may choose to interconnect further, requiring high-capacity HVDC to maintain their asynchronous attributes. For other areas around the globe where the "mesh" may not be as "dense," HVDC has and will have greater penetrations to connect new resources to load centers, to interconnect strong national and regional grids, and to connect offshore resources.
In the micro-world however, DC has made significant inroads in the U.S. and around the globe. End-use devices are ubiquitous these days with AC to DC conversions for each of them. This begs the question: Why endure the multiple conversions from distributed resources to end-use devices? Should we have separate DC networks built into new construction and retrofitted to accommodate more direct connections from distributed resources to end-use devices?
In summary, the macro-grid grows slowly and incrementally, so HVDC will have a slow but growing penetration. The real revolution is at the micro level with distributed resources and ubiquitous end-use devices. Will the next step be to lose the ubiquitous tethers, i.e., the multiple layers of spaghetti connected by hard wire to outlets?
With solar power generating DC electricity, it is obvious there would be benefits in avoiding the losses associated with converting to AC, especially when many electronic devices in the home run on DC anyway. Such benefits would be even further enhanced if DC could be used directly in lighting, charging electric vehicles and in appliances designed to run on DC, among other things.
To make this a reality, however, conversion services would have to be supplied by a utility over the grid to supplement and back up locally generated DC electricity. The big questions are can this be done economically, and if so, could it produce additional profits for utilities and manufacturers? If the answer to these two questions is yes, then where does the impetus come from to get things started? Are government subsidies needed or is there enough money in this to encourage industry to invest substantially?
Although signs of the journey back to DC are appearing, it will be interesting to see how it all unfolds and at what rate progress is achieved.
DC or AC – Can we all get along?
All electrical engineers and many others are very familiar with the historical significance of AC versus DC, the battle royal between Edison and Tesla/George Westinghouse. We also know that AC power won mainly because of the ability to transform voltages, which was necessary to transmit power over long distances and the need to reduce losses.
For a long time, it was certain that the battle and the war was over, and the future of the world was with AC. This is far from what is actually happening. The onset of electronics, PV, storage and others have brought a resurgence of DC-based supply and consumption. In addition, several advances in DC motors have made them almost as robust and efficient as AC motors.
This is causing a problem with the current AC grid because everything needs to be converted back into AC. Every electronic device in the house has a converter that converts the incoming AC to DC before the device consumes it. Most distributed energy sources also generate in DC, requiring converter mechanisms to convert them back into AC. This is not good. First of all, it is inefficient because all of these smaller converters are not designed to be energy efficient, and more importantly, there are just too many of them in a house.
We need to move to something that allows both to coexist. And here is a scenario that permits just that.
The future electric grid that ties the centralized generation sources to the load centers may still be AC-based, but not everything else needs to be AC. To start with, the houses can be pure DC-based. This is okay because much of the house works off of one single voltage level. Each house can contain one single converter to convert the incoming AC power to DC, and everything inside the house can be DC – right from the kitchen appliances to the electronics. In addition, the renewable sources of energy such as PV and other storage mechanisms can all continue to be DC-based and feed into the house. The main converter into the house can be two-way so that excess power can be converted back into AC and fed back into the grid. The physical separation between the two systems will also allow each homeowner to make their own energy decisions, and be better energy customers and better stewards of the environment as they feel appropriate. Lastly, reducing the number of converters in the house will also make the electronics simpler without the need for bulky and energy-inefficient conversion devices.
A future as described above allows us to create the perfect symbiotic relationship between the two systems by taking advantage of their respective strengths.
The electric power industry has historically remained largely unchanged since the pioneering days of Tesla. Yes, of course there have been major improvements in virtually all aspects of power delivery, including interconnected grids, insulation materials, transmission voltage levels and system efficiencies, but the basic principles are still in use today. The beauty of the AC system lies in the transformer and its ability to manipulate voltage levels throughout the process, from generator terminals to the outlets in your home. Based on the past history of relatively slow change in the industry, I do not foresee any rapid change on the horizon, at least in the area of generation and bulk power transmission within the confines of a mature U.S. market. While DC transmission may prove to be an overall better solution over an AC system in certain applications (sub-sea transmission, asynchronous AC interties and lower bulk transmission losses), the economic justification associated with modification or conversion from existing AC systems to DC technologies is generally not considered a viable option. I do think that we will see more use of DC technologies for transmission applications in the U.S. in the future as demand increases and existing resources reach capacity, but I would expect this to occur sporadically over quite a long period of time. Having said that, growing markets such as China, India and Africa are among a group of prime candidates for bulk DC transmission technologies. DC transmission is the optimal method for long-distance transport of bulk power.
With regards to the burgeoning demand caused by the massive influx of digital devices, such as computers, cell phones and tablets, there may be a compromise solution. I could foresee the use of add-on DC power distribution centers in a commercial facility, and perhaps even in the home, to create a type of “hybrid” energy design. I personally already make use of such devices, such as DC power “bricks,” to power dc electronics. However, I do not believe that it will go much farther than that. Without having the benefit of a study to draw reference from, I think it is a reasonable assumption to state that the costs associated with a commercial or residential retrofit from AC to DC would be prohibitive.
Utilities will have to adapt to accommodate new “green” technologies that continue to emerge into society. Elon Musk has announced his aggressive plans to build vast numbers of electric vehicles for the marketplace. As these vehicles become more commonplace, utilities will need to meet the demand by installing DC charging stations throughout their service territories. I do not foresee a change to DC grid distribution to handle this new need. While there are intrinsic losses relative to the use of electric vehicles, it remains doubtful that utilities will implement any type of DC transmission and distribution systems to support this endeavor. Existing generation facilities will not be impacted initially, but as demand continues to increase, so will the need for generation. Perhaps a fair amount of this energy will come from the modern-day renaissance of green generation sources such as wind and solar power, but except for special cases, I would expect that these will interconnect with the AC grid as they have been traditionally, for the foreseeable future.
HVDC has a very important role to play in transmission networks and is an economical viable option at ultra-high voltages (800 kV) in several regions in the USA. These DC links need to operate as hybrid HVAC and HVDC networks for integrating and controlling remote renewable resources, and they are well established to interconnect asynchronous areas. At the relative low capacity factors of wind (40%) and solar PV (20%) power production, HVAC networks are not utilized at an economic level for integrating these intermittent resources to the load centers, especially over long distances or with submarine cables from a relative weak remote AC network. HVDC is clearly the better option, especially the newer voltage source converter technologies. For no or very limited additional cost, two STATCOMs are added on both sides of the HVDC link. These STATCOMs are excellent to interconnect the remote power from relative weak networks to the load centers.
Multi-terminal HVDC networks and hybrid HVDC imbedded into HVAC networks are already commercialized and implemented in Europe and will provide the additional capacity to the transmission network in the USA. Furthermore these hybrid networks will at the same time increase the transmission capacity with reduced right-of-way (ROW) requirements. A HVDC link can transmit three to five times the power in the same ROW when compared to HVAC networks. At the same time, financing these new HVDC links is easier than traditional HVAC lines under merchant line agreements, since full control of the power flows can be guaranteed between remote resource and load center.
It is not a question of AC or DC but both, especially in the high-voltage and ultra-high-voltage transmission networks.
To join the DC power conversation, please comment in the box below.