OVER A THIRD OF POWER FROM NEW ENERGY – STUDY

NREL Study Shows Power Grid can Accommodate Large Increase in Wind and Solar Generation; Increased Coordination Over Wider Areas and More Frequent Scheduling Needed; Wind and Solar Significantly Reduce Carbon and Fuel Costs
May 20, 2010 (National Renewable Energy Laboratory)

THE POINT
It is a crucial, if geekish, point: Wind and solar power are NOT intermittent, they are VARIABLE.

If someone talks about problems with the intermittencies of wind and solar energies, it is out of ignorance or to intentionally discredit them. Obviously, the sun does not always shine and the wind does not always blow but those things can be scheduled and forecasted and are no reasons whatsoever not to keep building wind power and solar energy as fast as is humanly possible.

The Western Wind and Solar Integration Study, from the National Renewable Energy Laboratory (NREL) of the U.S. Department of Energy (DOE), says transmission system tools are now available that would allow the Mountain West and Southwestern states to get 35% of their electricity from wind and solar energies by 2017. All it will require, aside from the building of the wind and solar production capacity, is a change in the WestConnect group of grid operators’ standard operating procedures.

By adding more wind and photovoltaic and concentrating solar into the power supply, bringing the total for wind to 30% and for solar to 5%, electricity prices and greenhouse gas emissions (GhGs) can be reduced without a loss of reliability.

Electricity prices will fall as a result of fossil fuel cost savings of 40% and GhGs will likely fall 25-to-45%, the equivalent of taking 22-36 million cars off the road. For such extraordinary benefits, WestConnect power system operators would simply have to (1) expand the geographic area from which they draw, (2) make their electricity sales and purchases more frequently than the hour-by-hour schedule they presently use, (3) make more effective use of forecasting, and (4) increase the use of Demand Response (DR) programs that drive more efficient use of electricity when the need for it peaks.

click to enlarge

By expanding the geographic area from which grid operators draw solar and wind resources, they reduce variability because while it is true that the sun does not always shine and the wind does not always blow in any one place, it is almost always the case that either the sun is shining somewhere or the wind is blowing somewhere.

By scheduling power purchases/sales more frequently than every hour, grid operators reduce the need for “spinning reserves” and thereby eliminate some of the burning of natural gas and/or coal for back-up electricity that ultimately does not get used.

The more effective use of new, more accurate forecasting capabilities is another necessity for integrating more wind and solar and eliminating waste. It can further reduce costs 14%.

Utilizing DR programs to cut the need to supply more “peaking” electricity both eliminates the need for generation and the need for spinning reserves, making DR the cheapest option of all.

These shifts in grid operations will neutralize the issue of wind and solar energies’ variability and allow more than a third of the power supply to come from them without the loss of reliability or major new transmission undertakings.

click to enlarge

THE DETAILS
The Western Wind and Solar Integration Study (WWSIS) investigates the impact of getting up to 35% of grid electricity in the year 2017 from wind, photovoltaic solar (PV), and concentrating solar power (CSP) on the transmission and power generation system operated by the WestConnect group of utilities.

The WestConnect group of utilities serves Arizona, Colorado, Nevada, New Mexico, and Wyoming and includes Arizona Public Service, El Paso Electric Co., NV Energy, Public Service of New Mexico, Salt River Project, Tri-State Generation and Transmission Cooperative, Tucson Electric Power, Western Area Power Administration, and Xcel Energy.

The study took 2 and one-half years. It assumed larger balancing regions, more frequent scheduling, better use of transmission, state-of-the-art wind and solar forecasting, more flexible dispatching of generation, more solar and wind production, new transmission to make more solar and wind available to the grid, and new and more fully-used existing Demand Response (DR) programs.

click to enlarge

The WWSIS and the previously completed Eastern Wind Integration and Transmission Study (EWITS) (see NREL STUDY SAYS NEW ENERGY FITS FINE) came out of the U.S. Department of Energy (DOE) 20% Wind Energy by 2030 Study that showed it is feasible for the U.S. to obtain that amount of its power from wind by that year.

Solar power was included in WWSIS because of the immense solar resources and development in the West. Four of the five WestConnect states have a Renewable Eectricity Standard (RES) requiring that 15-to-30% their power come from New Energy sources by 2020 or 2025.

click to enlarge

The WWSIS also considered the entire western transmission system, called the Western Interconnection, and the impact as much as 23% wind/solar on the Western Electricity Coordinating Council (WECC) covering the entirety of the western states.

The questions asked by the WWSIS:
(1) What is the operating impact of carrying up to 35% New Energy and how can this be managed?
(2) How does a larger geographic region help to offset variability?
(3) How do local New Energy resources compare to more distant but higher quality New Energy resources delivered by long distance transmission?
(4) Can cooperation between power systems over larger areas reduce variability?
(5) How should reserve requirements be managed to best deal with the variability in wind and solar?
(6) What benefits come with using wind and solar forecasting in grid operations?
(7) How can hydro generation help make wind and solar more viable?

click to enlarge

The study is the completion of preliminary work and the EWITS. It was funded by DOE. The wind power dataset, and the wind and solar forecasts were provided by the 3TIER Group. There was a variety of academic and industry contributions and a Technical Review Committee (TRC).

The WWSIS used historical load and weather patterns from years 2004, 2005, and 2006 to make its predictions. It assumed a wind capital cost of $2000/kW (2008$), PV at $4000/kW, CSP with thermal storage at $4000/kW, transmission at $1600/MW-mile, and transmission losses at 1% per 100 miles. It assumed no production or investment tax credit benefits. It assumed existing transmission to be unavailable for remote solar and wind energy resources and that 0.7 MW of new transmission would be added for each 1.0 MW of new wind or solar generation.

Assumed costs: 2017 nominal dollars and a 2% per year increase in prices. Coal: $2/MBTU; Natural gas: either $3.50/MBTU or $9.50/MBTU; Carbon dioxide (CO2): $30/metric ton. At the higher price, the use of natural gas is displaced by wind and solar; at the lower price, they displace coal.

click to enlarge

A reasonable set of assumptions are specified for balancing, reserves, O&M costs and hydro costs. The study also specifies that it is an operations study, not a cost-benefit analysis, a transmission planning study or a reliability study. It does not address stability or the ideal balance between wind and solar.

The study incorporated data from 960+ GW of wind sites, 15+ GW of PV plants and 200+ GW of CSP plants (with thermal storage).

Four levels of wind and solar energy were studied.
Case 1: The Preselected case is based on the 2008 wind and solar capacity.
Case 2: The 10% case is based on 10% wind energy and 1% solar energy (70% CSP and 30% PV).
Case 3: The 20% case is based on 20% wind energy and 3% solar energy in the WestConnect and 10% wind energy and 1% solar energy in the rest of WECC.
Case 4: The 20/20% case is based on 20% wind energy and 3% solar energy in the WestConnect and in the WECC.
Cae 5: The 30% case is based on 30% wind energy and 5% solar energy in the WestConnect and 20% wind energy and 3% solar energy in the rest of WECC.

click to enlarge

Three geographic scenarios were considered, balancing between wind and solar close to the load (demand) but not in great quantities and large supplies of wind and solar distant from the load.
(1) The In Area Scenario obtains all wind and solar energy within state boundaries and adds no new interstate transmission.
(2) The Local Priority Scenario uses wind and solar in the entire WestConnect region but cuts 10% from the cost for in-state resources and adds little transmission.
(3) The Mega Project Scenario uses the best available wind and solar in the WestConnect, much in Wyoming, and adds extensive new transmission lines to deliver it.

click to enlarge

To add 35% New Energy, variability must be managed.

All three geographic scenarios show significant benefits with no negative effects in the 10% case.

No significant adverse impacts were observed up to the 20% case in WestConnect.

Increased New Energy in the rest of WECC (20/20% case) led to stress on system operations in the WestConnect and some instances of insufficient reserves due to wind and solar forecast errors.

click to enlarge

Managing variability is more challenging in the 30% case and load and contingency reserves can be met only if wind and solar forecasts are perfect. If forecasts fail, load can be met but there would be reserve shortfalls. Extra spinning reserves can be held every hour of the year to meet those shortfalls but the cost would be very high. Adding a demand response (DR) program or developing better forecasts are the best strategies to make the 30% case work.

Wind and solar replace natural gas at the higher projected costs. Because gas is more flexible than coal, the system becomes less flexible but operating costs in the 30% case drop by $20 billion/yr, estimated at $80-to-$88/MWh, from ~$50 billion/yr, a 40% savings in fuel and GhG costs. The costs of building the wind, solar and transmission infrastructure needed to have access to a wider area and implement sub-hourly scheduling could come from the system’s cost savings.

click to enlarge

If gas costs $3.50/MBTU, wind and solar replace coal and the system is more flexible. Operating costs are cut ~40% ($46/MWh) but GhG cuts are greater, adding to savings.

There key benefits of a wider balancing area: (1) More New Energy, offsetting variability, (2) reduced overall variability of the load, and (3) more non-New Energy balancing resources can be pooled, adding the availability of more flexible (natural gas) balancing resources. $2 billion could be saved by using the 106 zones of the WECC operating costs in the 10% case.

Current practice is to schedule generation and interstate exchange once per hour. This can make it difficult to manage sub-hourly variability. Minute-to-minute simulations show hourly scheduling impacts regulation even more than the variability of wind and solar. In the 30% case, fast maneuvering with sub-hourly scheduling costs ~50% of hourly scheduling. Sub-hourly scheduling also improves plant efficiency and cuts O&M costs.

click to enlarge

Day-ahead wind and solar forecasts help mitigate the wind and solar variability. generation. Forecasts are imperfect. Wrong forecasts cause reserve shortfalls. Over a year, forecasts reduce WECC operating costs by up to 14%, which is $5 billion/yr or $12-to-$20/MWh compared to ignoring day-ahead forecasts. Perfect wind and solar day-ahead forecasts would cut WECC operating costs another $500 million/yr in the 30% case, which is $1-to-$2/ MWh.

On average, wind forecast error is not very large (8% across WestConnect). But extreme forecast errors can be costly. Spinning reserves can be increased to cover such events but instead of holding additional spinning reserves for each of the 8760 hours of the year, a robust DR program can meet the shortfall at a much lower cost.

click to enlarge

Wind curtailment can also occur, most often when load falls, wind rises and spinning reserves are inflexible (with coal at more than 70%). It happens for approximately 0.5% of all wind generation.

Utilities are required to cover 10-minute load variability 95% of the time. Typically, the existing dispatchable generating fleet can meet the requirement. Regulating reserves, automatically controlled, are part of dealing with variability. Wind is useful as a regulating reserve beause it is easily curtailed but with an adequately widespread balancing area, WestConnect can handle large amounts of wind and solar, and curtailment of wind is expected to be no more than 1% or less in the 30% case.

click to enlarge

The In-Area, Local Priority, and Mega Project Scenarios showed similar performances and comparable economics at each level of wind and solar. This is further evidence that new long distance high voltage transmission is less necessary than lines that get new wind and solar to the existing lines.

Storage can provide many benefits, not least of which is lowering costs by providing electricity generated during off peak hours to meet peak demand. The leading forms of storage are pumped storage hydro (PSH), solar thermal storage, and plug-in hybrid electric vehicles (PHEVs). None is currently an economically viable investment.

A best-case scenario for storage would be a new 100-MW PSH plant. Even with perfect foresight of spot prices, it would have an operating value of about $2.6 million per year without wind and solar. With 30% wind and solar, a new 100-MW PSH plant would have an operating value of $0.5 million per year, $35/kW. System flexibility is much more valuable than storage.

Including hydro generation, which is capable of quick start/stop cycling and fast ramping and therefore a good match with variable generation, cuts cost.

click to enlarge

Reducing coal plant output to 40% of nameplate capacity, as is assumed in the WWSIS, increases the operating costs of a coal plant. In the 30% scenario, coal plants operate above 70% load and WECC operating costs increase ~$160 million/yr.

Variable resources are considered “energy” resources and not “capacity” resources. They add to reliability but have a range of capacity. In the WWSIS study, wind had capacity values in the range of 10% to 15% and was higher during winter and spring and at night. PV solar plants have capacity values in the range of 25% to 30%, is highest during the daytime and declines in the late afternoon and early evening before peak load falls off. Concentrating solar plants with thermal energy storage have capacity values in the range of 90% to 95%, similar to conventional thermal generating plants. Their maximum energy production is during long summer days. Storage capability extends solar power plant energy output through peak demand.

click to enlarge

QUOTES
- From the WWSIS study conclusion: “The technical analysis performed in this study shows that it is feasible for the WestConnect region to accommodate 30% wind and 5% solar energy penetration. This requires key changes to current practice, including substantial balancing area cooperation, sub-hourly scheduling, and access to underutilized transmission capacity.”

click to enlarge

- From the WWSIS study: “WWSIS finds that both variability and uncertainty of wind and solar generation impacts grid operations. However, the uncertainty (due to imperfect forecasts) leads to a greater impact on operations and results in some contingency reserve shortfalls and some curtailment, both of which are relatively small. The variability leads to a greater sub-hourly variability reserve requirement, but because conventional units are backed down, the system naturally has extra reserve margins.”

click to enlarge

- From the WWSIS study conclusion: “This study has established both the potential and the challenges of large scale integration of wind and solar generation in WestConnect and, more broadly, in WECC. However, changes of this magnitude warrant further investigation. The project team regards the following as valuable topics for exploration: • Characterization of the capabilities of the non-renewable generation portfolio in greater detail (e.g., minimum turndown, ramp rates, cost of additional wear and tear); • Changes in non-renewable generation portfolio (e.g., impact of retirements, characteristics, and value of possible fleet additions or upgrades); • Reserve requirements and strategies (e.g., off-line reserves, reserves from non generation resources); • Load participation or demand response (e.g., functionality, market structures, PHEV); • Fuel sensitivities (e.g., price, carbon taxes, gas contracts and storage, hydro constraints and strategies); • Forecasting (e.g., calibration of forecasting using field experience, strategies for use of short-term forecasting); • Rolling unit commitment (e.g., scheduling units more frequently than once on a day-ahead basis); • Transmission planning and reliability analyses (e.g., transient stability, voltage stability, protection and control, intra-area constraints and challenges); • Hydro flexibility (e.g., calibration of hydro models with plant performance).”