Costs for utility-scale solar photovoltaic (PV) systems have declined in recent years—most sources show that system costs on a per-watt basis have fallen about 10% to 15% per year from 2010 through 2016. The level of those costs in certain years often varies across sources for reasons largely attributable to the way these costs are estimated.
To estimate capital costs of generating technologies, analysts use one of two common methods—total reported costs or aggregated component costs. Both approaches help explain the cost of utility-scale solar PV systems.
Reported costs: Using actual project data provides an empirical analysis that captures a large range of reported project costs in the market and accounts for the substantial variability in project design, location, and timing observed in the real world. Challenges with this approach include uncertainty about whether certain cost components are included in reported system costs, such as interconnection costs and the treatment of financing expense. Also, the data for each year reflect projects completed in that year, which do not necessarily reflect the costs of projects initiated in that year.
Component costs: The component cost approach provides more detail on the impact of changes in component-level technology and costs, which can be significant in a fast-moving market like solar PV. Such approaches typically represent either best-in-class or common-practice project criteria and do not necessarily capture the wide range of real-world project cost factors. Estimates that exclude financing expenses are called overnight estimates (i.e., as if the plant could be built instantly with no financing requirement). Component-based estimates may not reflect all potential costs to a system, such as developer profit margins.
EIA started collecting data on total capital costs directly from project owners as a part of the Form EIA-860 Annual Electric Generators Report in 2013. Because of respondent confidentiality, EIA only publishes capacity-weighted average values of new projects coming online each year and has published data for 2013, 2014, and 2015. This data series includes facilities with a nameplate capacity of at least one megawatt of alternating current. Respondents are asked to exclude government incentives and financing expenses from the reported costs.
The U.S. Department of Energy’s Lawrence Berkeley National Laboratory (LBNL) begins with EIA’s capital cost dataset and gathers additional information from corporate financial reports, Federal Energy Regulatory Commission (FERC) filings, and the U.S. Department of the Treasury’s Section 1603 grant database. LBNL’s annual Utility-Scale Solar Report defines utility-scale solar facilities as those with at least five megawatts or more of alternating current, which cuts out some of the smaller plants included in EIA’s Electric Generator Report.
The U.S. Department of Energy’s National Renewable Energy Laboratory (NREL) publishes the Solar PV System Cost Benchmark report with estimates of total system costs based on the most up-to-date information on reported component costs and conversations with industry. These costs do not include additional net profit components, which are common in the marketplace. Also, NREL’s bottom-up approach models costs for a project sized at 100 megawatts of direct current, which is large enough to have realized some economies of scale relative to smaller systems.
EIA also projects future capital costs as part of the Annual Energy Outlook (AEO). Starting costs of solar PV come from contracted capital cost studies based on information on system design, configuration, and construction derived from actual or planned projects, using generic assumptions for labor and materials rates.
Although EIA does not update the capital cost study each year, in years where the report data are not updated, EIA extrapolates cost trends observed in the literature, including the sources noted above, and considers expected cost declines from learning-by-doing. For 2018, AEO2018 projects installed capital costs of $1.85 per watt (AC) for fixed-tilt PV systems and $2.11 per watt (AC) for single-axis tracking systems.
Principal contributor: Cara Marcy
Something interesting to share?
Join NrgEdge and create your own NrgBuzz today
Less than two weeks ago, the VLCC Navarin arrived at Tanjung Pengerang, at the southern end of Peninsular Malaysia. It was carrying two million barrels of crude oil, split equally between Saudi Arab Medium and Iraqi Basra Light grades.
The RAPID refinery in Johor. An equal joint partnership between Malaysia’s Petronas and Saudi Aramco whose 300 kb/d mega refinery is nearing completion. Once questioned for its economic viability, RAPID is now scheduled to start up in early 2019, entering a market that is still booming and in demand of the higher quality, Euro IV and Euro V level fuels RAPID will produce.
Beyond fuel products, RAPID will also have massive petrochemical capacity. Meant to come on online at a later date, RAPID will have a collective capacity of some 7.7 million tons per annum of differentiated and specialty chemicals, including 3 mtpa of propylene. To be completed in stages, Petronas nonetheless projects that it will add some 3.3 million tons of petrochemicals to the Asia market by the end of next year. That’s blockbuster numbers, and it will elevate Petronas’ stature in downstream, bringing more international appeal to a refining network previously focused mainly on Malaysia. For its partner Saudi Aramco, RAPID is part of a multi-pronged strategy of investing mega refineries in key parts of the world, to diversify its business and ensure demand for its crude flows as it edges towards an IPO.
RAPID won’t be alone. Vietnam’s second refinery – the 200 kb/d Nghi Son – has finally started up this year after multiple delays. And in the same timeframe as RAPID, the Zhejiang refinery by Rongsheng Petro Chemical and the Dalian refinery by Hengli Petrochemical in China are both due to start up. At 400 kb/d each, that could add 1.1 mmb/d of new refining capacity in Asia within 1H19. And there’s more coming. Hengli’s Pulau Muara Besar project in Brunei is also aiming for a 2019 start, potentially adding another 175 kb/d of capacity. And just like RAPID, each of these new or recent projects has substantial petrochemical capacity planned.
That’s okay for now, since demand remains strong. But the danger is that this could all unravel. With American sanctions on Iran due to kick in November, even existing refineries are fleeing from contributing to Tehran in favour of other crude grades. The new refineries will be entering a tight market that could become even tighter. RAPID can rely on Saudi Arabia and Nghi Son can depend on Kuwait, both the Chinese projects are having to scramble to find alternate supplies for their designed diet of heavy sour crude. This race to find supplies has already sent Brent prices to four-year highs, and most in the industry are already predicting that crude oil prices will rise to US$100/b by the year’s end. At prices like this, demand destruction begins and the current massive growth – fuelled by cheap oil prices – could come to an end. The market can rapidly change again, and by the end of this decade, Asia could be swirling with far more oil products that it can handle.
Upcoming and recent Asia refineries:
Headline crude prices for the week beginning 8 October 2018 – Brent: US$84/b; WTI: US$74/b
Headlines of the week
Source: U.S. Energy Information Administration, Monthly Crude Oil and Natural Gas Production
As domestic production continues to increase, the average density of crude oil produced in the United States continues to become lighter. The average API gravity—a measure of a crude oil’s density where higher numbers mean lower density—of U.S. crude oil increased in 2017 and through the first six months of 2018. Crude oil production with an API gravity greater than 40 degrees grew by 310,000 barrels per day (b/d) to more than 4.6 million b/d in 2017. This increase represents 53% of total Lower 48 production in 2017, an increase from 50% in 2015, the earliest year for which EIA has oil production data by API gravity.
API gravity is measured as the inverse of the density of a petroleum liquid relative to water. The higher the API gravity, the lower the density of the petroleum liquid, meaning lighter oils have higher API gravities. The increase in light crude oil production is the result of the growth in crude oil production from tight formations enabled by improvements in horizontal drilling and hydraulic fracturing.
Along with sulfur content, API gravity determines the type of processing needed to refine crude oil into fuel and other petroleum products, all of which factor into refineries’ profits. Overall U.S. refining capacity is geared toward a diverse range of crude oil inputs, so it can be uneconomic to run some refineries solely on light crude oil. Conversely, it is impossible to run some refineries on heavy crude oil without producing significant quantities of low-valued heavy products such as residual fuel.
Source: U.S. Energy Information Administration, Monthly Crude Oil and Natural Gas Production
API gravity can differ greatly by production area. For example, oil produced in Texas—the largest crude oil-producing state—has a relatively broad distribution of API gravities with most production ranging from 30 to 50 degrees API. However, crude oil with API gravity of 40 to 50 degrees accounted for the largest share of Texas production, at 55%, in 2017. This category was also the fastest growing, reaching 1.9 million b/d, driven by increasing production in the tight oil plays of the Permian and Eagle Ford.
Oil produced in North Dakota’s Bakken formation also tends to be less dense and lighter. About 90% of North Dakota’s 2017 crude oil production had an API gravity of 40 to 50 degrees. The oil coming from the Federal Gulf of Mexico (GOM) tends to be more dense and heavier. More than 34% of the crude oil produced in the GOM in 2017 had an API gravity of lower than 30 degrees and 65% had an API gravity of 30 to 40 degrees.
In contrast to the increasing production of light crude oil in the United States, imported crude oil continues to be heavier. In 2017, 7.6 million b/d (96%) of imported crude oil had an API gravity of 40 or below, compared with 4.2 million b/d (48%) of domestic production.
EIA collects API gravity production data by state in the monthly crude oil and natural gas production report as well as crude oil quality by company level imports to better inform analysis of refinery inputs and utilization, crude oil trade, and regional crude oil pricing. API gravity is also projected to continue changing: EIA’s Annual Energy Outlook 2018 Reference case projects that U.S. oil production from tight formations will continue to increase in the coming decades.