New analytical tool measures generation costs on price of natural gas, strength of the wind and sun
Energize Weekly, July 12, 2017
Utilities across the country, when building new generation, are opting for natural gas-fired plants, wind farms or solar installations. The three have dominated industry investment and a new analytical tool could aid utility executives in deciding which to build.
The electricity industry is already planning to increase natural gas-fired generating capacity by 36.6 gigawatts in the next two years, as well as adding 21 gigawatts of wind and 11 gigawatts of solar, according to the U.S. Energy Information Administration.
The new approach devised by Lawrence Livermore National Laboratory researcher Mark Bolinger focuses on a key risk and variation that accompany each of the types of generation – the price of natural gas and the strength of the wind or sun.
The standard way of comparing generating sources is through “levelized cost of energy” (LCOE) studies, which calculates the cost of building and operating a plant over its lifetime and divides that by the amount of energy it generates giving a dollars per megawatt-hour price.
Bolinger takes the LCOE a step further by introducing “resource” risk for each of type of generation in his study “Using Probability of Exceedance to Compare the Resource Risk of Renewable and Gas-Fired Generation.”
For renewable energy generation the resource risk is a “quantity risk” in that the amount of wind or insolation could be lower than expected. For natural gas-fired units the risk is in the price – higher than expected fuel costs.
To measure these risks Bolinger constructed statistical models that over a 25-year period looked at the range of prices for natural gas from low to high and variations in wind and solar resources. These were then used to generate a range of LCOEs.
“Most often, resource risk—and natural gas price risk in particular—falls disproportionately on utility ratepayers, who are typically not well-equipped to manage this risk,” Bolinger wrote. “As such, it is incumbent upon utilities, regulators, and policymakers to ensure that resource risk is taken into consideration when making or approving resource decisions.”
The risks posed by the variations in wind and sun or in the prices of natural gas are very different. In Bolinger’s models those variations are ranged around an average, or mid-range, case labeled P50. The variations run from P1 to P99.
The renewable resource risk is symmetrical about P50 and over time declines as the wind or solar resource is better characterized. The natural gas risk is asymmetrical – skewed towards higher prices – and increases over time, as it is harder to project what will happen in natural gas markets.
“In general, higher-than-expected gas prices appear to be riskier to ratepayers than lower-than-expected wind or solar output,” Bolinger said. “This suggests that from a ratepayer perspective, we should perhaps be more concerned about gas price risk than about wind or solar resource risk.”
Historically, natural gas prices have been very volatile, but the discovery of large shale gas reserves in the U.S. in the last decade has muted that volatility and kept prices low. Bolinger uses the price variations since 2008, when shale gas started to come on the market and prices have been low.
This makes natural gas more competitive, but the analysis notes that while this is the most “logical choice” it assumes prices and volatility will be kept in check over the next 25 years.
Even with the lower volatility in natural gas prices, there is a growing market for the price stability offered by low-cost wind and solar projects, the study said. Wind and solar are sold on long-term, fixed contracts, while natural gas generation is sold in spot markets and on short-term or “tolling contracts” in which fuel costs are passed along the power purchaser.
So how do wind and solar stack up with natural gas in Bolinger’s models?
Wind, without the production tax credit, is more expensive than natural gas over the next 24 years at P50 (the cost curves meet in the 25th year). But at P25 wind is cheaper than all natural gas-fired variations in 18 years and at P1 wind beats natural gas after two years.
The story is similar for solar. At P50 natural gas-fired generation is cheaper than utility-scale solar, even with the 30 percent investment tax credit the projects are granted, but as natural gas prices rise the solar plants become more competitive over time.
“This framework demonstrates that while gas-fired generation is competitive with or cheaper than wind and solar generation under modeled conditions on an expected or “P50” basis, worse-than-expected scenarios often yield the opposite conclusion—i.e., that wind and solar power are cheaper than gas-fired generation,” the report concludes.
Looking at resource risk is, however, just part of the equation and LCOE is only one measure in the decision making on infrastructure investment, Bolinger notes. “In this sense, it should be recognized from the start that this report is focused on just one side of a two-sided coin,” the report said.