by Eric B. ForsythRev 11/22/11 Everyone knows that it is dangerous to make predictions, we live in a world of random happenstance: wars, plagues, natural disasters, etc. But there is one prediction we can make about the electric power system with some certainty: when the fossil fuel reserves are gone it will be completely different from what we have now. That’s because at present about 66% of the power generated in the US is derived from plants fired by coal or natural gas, in passing note that less than 1% of electricity is produced by burning oil. Green enthusiasts promote replacement by renewable sources such as wind turbines and photo-voltaics, but there is a problem that becomes increasingly onerous if these sources of energy are to provide a dominant percentage of generation, namely the primary driving energy is intermittent. There are windless days and, of course, the sun is absent on average for half the day or more. Data from large wind installations in Europe show that the ratio of average to maximum power is about 25 % over a year (this ratio is termed the capacity factor or CF). Solar generators are not much better although there is much less data and performance is very location sensitive. Because at present power generated by both these renewables is very small compared to the rest of the system, fluctuations caused by their intermittent nature can be absorbed without major instability problems. However utilities are installing gas turbines because it is anticipated that steam generators will not respond fast enough as the percentage of renewables increases.

SYSTEM REQUIREMENTS

Once fossil-fuel plants are history this means intermittent renewable energy cannot stand alone; generation must be one leg of a three-legged stool. To provide high quality, i.e. uninterrupted, electricity, two other legs are needed: a flexible interconnected transmission system and massive energy storage with a fast response time. The flexible transmission system will be needed because the availability of both forms of renewable energy can vary widely from one geographic area to another and with time. A flexible transmission system is conceptually not too difficult to envisage, although it may be expensive. Massive energy storage is another story, at present this usually accomplished by raising the potential energy of water and generating electricity by conventional hydro-electric plants. Such favorable geographical sites which are not already developed do not occur in much of the US. It is instructive to estimate the amount of energy storage needed to stabilize the entire country in the absence of fossil fuel plants. Although average values for the entire country are used in this article, in practice the country would have to be divided into zones, similar to regions under the control of the North American Electric Reliability Corporation, which was formed to prevent black-outs. And, of course, the ‘average’ is the average of fluctuating peaks and valleys, the system must have margins to meet the peaks, day and night, winter and summer. There is no cut and dried formula to calculate the stored energy needed to stabilize a zone. At any time, the amount depends of the distribution of the renewable energy, such as wind, across the zone and the availability of interconnections. Judging the amount of energy stored to keep the network stable and running at a constant frequency is an exercise in probability, so that the chance of running out of the budgeted amount of stored energy is vanishingly small.

Current annual power generation in the US is about 4000×10³ GWh. If the population grows at 1% per year and per capita electrical consumption remains the same this extrapolates to about 6600x 10³ GWh per year in 50 years or 18×10³ GWh/day. Of course in the fifty year period the nature of the demand will change; electrical heating will replace oil and gas fired heating for example, but at the same time conservation and improved efficiency would reduce demand, assuming these factors tend to cancel each other, 18×10³ GWh/day is a reasonable starting point.

THE INTRODUCTION OF ELECTRIC VEHICLES

America is a mobile society and automobiles are at the core of our economic system. The production, maintenance and powering of cars accounts for a good deal of domestic economic activity. If electrically powered vehicles replace our present fleet of fossil-fueled cars, this change will significantly impact electrical demand on the grid. For this analysis trucks and buses are not included as they might well be powered differently using a fuel derived from biomass. The introduction of electric cars has scarcely begun in 2011 and performance is limited, battery capacity of 15 to 100 kWh per vehicle restricts the range. However, battery design has greatly improved over the past decade and it is reasonable to assume that on-board energy storage will rise to about 200 kWh or more in the time frame under consideration, this would provide 5 to 8 hours of highway driving, with fast recharge an option after that.

At present the average automobile usually carries about 50% of its fuel capacity at any one time and the range is typically 250 miles from a full tank, although according to Dept of Transportation statistics the average mileage driven is only 36 miles/day. The availability of massive amounts of energy storage in the propulsion batteries of purely electric vehicles raises the possibility of using some of the energy to stabilize the grid, provided cars are ‘plugged in’ when stationery. The idea first emerged more than ten years ago and there is considerable literature on the subject, although in the past battery performance was too poor to encourage large practical tests, or even merit serious consideration of the scheme. Computer simulations are encouraging if battery performance continues to improve, the topic is often referred to by the acronym V2G (Vehicle to Grid). The advantages can be summarized:

• The batteries are needed anyway for vehicle propulsion.
• The batteries store much more energy than needed for the typical daily use of the car.
• The distribution and ultimately the transmission system to which the converters are connected must be adequately sized to permit fast recharging if needed, this would also allow significant energy flow back into the network.
• Batteries and converters will be widely distributed throughout the load area.
• Modern solid state converters are capable of fast response and easily controlled remotely.
• The concept matches the future expectation of increased use of battery-powered vehicles and more use of renewable intermittent generation. Over time, fossil-fueled vehicles may well be phased out at about the same time as fossil-fueled power plants. Coordinating the use of electrically powered vehicles and a significant contribution to the grid by intermittent sources could provide a major step to the goal of a future energy balanced ‘green’ society.

NUMERICAL EXAMPLE

If vehicle ownership tracks population growth we can expect about 410 million private passenger cars on the road in 50 years. Assuming 200 kWh per vehicle, this corresponds to 82×10³ GWh total when all the batteries are charged. As mentioned, daily use is much less than the maximum possible, this is a crucial point in determining how much can be siphoned off for network stabilization. The ratio of daily use to maximum capacity is the demand factor, or DF.

For a DF of 12% the energy drawn from the grid for recharging, assuming the losses are low, is 0.12 x 82×10³ GWh per day, or 9.8x 10³ GWh.

This demand must be added to the conventional load calculated above of 18x 10³ GWh per day to yield an average total of 27.8x 10³ GWh per day for the US as a whole.

The next question is to get a feel for how much of this daily load could be supported by intermittent sources such as wind turbines and photo-voltaic cells.

NETWORK STABILIZATION

The reserve factor, or RF, is defined as the ratio of energy that could be supplied to the network compared to the maximum energy stored without jeopardizing seriously the recharging of each vehicle. For example an RF of 15% under the worst conditions would mean there is always a minimum of 85% available for vehicles, which should be adequate with a DF of 12%. The converter at each charging point, which either supplies direct current for charging or alternating current for grid stabilization would be controlled either locally, by the car owner, or remotely from a regional control center. The owner would have the choice of fast or normal ( typically 8 hours) charging, probably under the supervision of a ‘smart’ grid, and the remote controller would determine how much to draw down the battery when needed to compensate for lack of power from intermittent sources. The converter could even be ‘intelligent’, and adjust for the driving habits of each owner. The regional controller for each zone would determine how much and when energy is returned to grid from the vehicle batteries to stabilize the network. The infrastructure of batteries and converters is ideal for this purpose because it is widely distributed, already wired for sufficient power transfer and capable of very fast response time.

In the numerical example a RF of 15% would yield about 12×10³ GWh/day for stabilization purposes. Just how much intermittent power could be safely operated without threatening network instability with this much stored energy available depends on the reserve margins. It might even depend of wind or sunlight forecasts. But on average it could be assumed the contribution from intermittent sources will match the RF of 15%, or in this example about 12×10³ GWh/day, which corresponds to 12/27.8 of total demand or 43%. This suggests intermittent sources could provide slightly more than the increased demand caused by the introduction of electric cars, i.e, 12 compared to 9.8( x10³ GWh/day)

RENWABLE ENERGY REQUIRED

Assuming that the energy is provided by wind turbines we can figure the number and extent of the system. Of course, other intermittent sources could also contribute, such as photovoltaic cells, and other storage methods, such as dams, could be mixed in the overall system. Assuming wind turbines have a capacity factor of about 25%, a daily output of 12×10³ GWh would require an installed maximum capacity of 48×10³ GWh/day. If the individual turbines are rated at 3.5 MW then about 570,000 units are needed. If installed in parks of 100 units, each park would cover about 5 square miles and the total for the US as a whole equals 28,500 square miles or an area slightly larger than West Virginia. Of course, offshore parks could also be built. The average generating capacity of a park, about 88 MW, corresponds to a small fossil fuel plant. The direct cost, not including interconnections to substations and real estate, would be about $2 trillion, or $40billion/year over 50 years on average (present value).

THE BALANCE OF ELECTRIC GENERATION

If intermittent sources provide about 12×10³ GWh/day, what about the balance of 27.8 -12 or 15.8×10³ GWh/day? As with the present system a small percentage could be provided by hydroelectric, geothermal and cogeneration. It is possible that a fusion reactor may be on line in 50 years. However even an optimistic guess would not put these sources at more than 10% of the balance needed, say 1.6×10³ GWh/day, leaving 14.2×10³ GWh/day. The only source capable of generating power of this scale is nuclear fission, possibly using advanced designs such as breeder or recycling reactors. The recent tragedy in Japan will probably lead to design improvements and stricter siting criteria. Unfortunately technical progress is often made with feedback from accidents, the airline transportation business is a case in point, which is now very safe despite numerous accidents in the past. If the balance is provided by nuclear reactors rated at 1 GW with a CF of 0.9, then about 660 plants will be needed, without accounting for extra capacity to provide reserve power.

CONCLUSION

As fossil fuel plants are phased out, intermittent renewable energy sources can only be effectively utilized in combination with a flexible transmission and distribution network and widely distributed energy storage. In the far future the most versatile form of energy storage will be propulsion batteries for automobiles. Using some of this stored energy to stabilize the network in combination with fluctuating, intermittent generation from wind turbines and solar panels could maximize the number that can be brought on line. But it seems unlikely that intermittent renewables will provide more than half the energy needed in the face of population growth and changing uses of electricity. Experience with an actual operating system may permit refinement of DF and RF to safely maximize the contribution of renewable energy. The only viable source for the bulk of the balance is nuclear fission reactors, possible of advanced designs, unless the R&D on fusion reactors reaches fruition in the far future.

The wide-scale introduction of vehicles propelled purely by rechargeable batteries will put the electric utilities in competition with the oil companies. This development will produce a host of political problems during the transition period as electric vehicles are phased in. Each battery will have to be electronically identified to ensure proper billing for power used for charging and proper credit for power withdrawn for grid stabilization. This will require high levels of security and privacy protection. The car-owning public will have to agree to permit their vehicle to participate in this kind of system but a financial incentive could be devised. If battery improvement continues at the present rate there is a good chance the driving public will switch to electrically-powered cars before fossil-fueled power plants disappear, such cars will potentially require less maintenance and provide performance on the road equal or better than present-day automobiles.

The numerical examples are speculative but they illustrate several important points:

• Population growth will place a severe burden on the grid if per capita consumption remains the same.
• If battery performance continues to improve, electrically-powered vehicles may actually facilitate the incorporation of wide-scale intermittent ‘green’ energy, perhaps as much as 40 to 50 % of the average power supplied by the grid.
• Massive contribution to the grid by intermittent sources will consume a huge amount of real estate, although limited agriculture may be possible on the land.
• On the scale predicted a significant amount of generation must come from nuclear reactors, although design and siting criteria must be changed.

Brief CV of Eric B. Forsyth Mr. Forsyth grew up in England where he served as an RAF fighter pilot in the 1950s. He obtained a master’s degree at Toronto University in 1960 and then worked until his retirement in 1995 at Brookhaven National Laboratory on Long Island, NY. He led the development at Brookhaven of superconducting cables suitable for very high capacity underground ac transmission systems. In 1986 he was appointed chair of the Accelerator Development Department which was responsible for the construction and design of several particle accelerators including preconstruction design and planning of the Relativistic Heavy Ion Collider, now the largest nuclear physics research tool in the US. Since retirement he has taken his sailboat twice round the world and sailed to both Polar Regions several times including a transit of the Northwest Passage. He has observed first-hand the effects of climate change and the efforts in many countries to deal with the future energy crisis. He is a Fellow of IEEE and in 2007 he was presented with the Herman Halperin Award for Transmission and Distribution development.