TODAY’S STUDY: WIRES TO CONNECT THE HIGH PLAINS WITH THE WEST COAST
California-Wyoming Grid Integration Study; Phase 1—Economic Analysis
D. Corbus, D. Hurlbut, P. Schwabe, E. Ibanez, M. Milligan, G. Brinkman, A. Paduru, V. Diakov, and M. Hand
March 2014 (National Renewable Energy Laboratory)
This report examines the economics of new transmission to deliver wind power from Wyoming to electricity customers in California. It looks at possible choices for meeting the last increment of California’s renewables portfolio standard (RPS) requirement—33% of retail sales by 2020— comparing Wyoming wind with in-state renewables likely to be available in 2017. Other recent studies take a system-wide approach to renewable energy expansion in California and the West.1
The purpose of this study is to match these system-wide analyses with one that focuses on a possible Wyoming-to-California transmission corridor, which a number of the regional studies suggest could accommodate cost-effective renewable energy delivery. By focusing on the Wyoming-California transmission corridor, this analysis replaces system-wide assumptions with inputs that are more specific, providing a more fine-tuned analysis of costs and benefits and enabling a more detailed assessment of risk.
The results suggest that the economic benefits of developing the corridor exceed the costs under the array of future conditions tested in the analysis. Benefit-to-cost ratios range from 1.62 to 3.62 depending on assumptions about federal tax incentives in 2017, and depending on assumptions about the future costs of different renewable energy technologies. Where outcomes fall within this range will depend on:
• Expectations about future technology costs. If large-scale solar photovoltaic (PV) costs fall significantly faster than the cost of wind power, the ratios will tend toward the lower end of the ranges reported here.
• Expectations about future federal tax incentives. Reductions in the production tax credit (PTC) and investment tax credit (ITC) tend to favor developing the corridor, particularly if the reductions are even across all benefitting renewable technologies. If the changes significantly benefit solar and geothermal without benefitting wind, the ratios will tend toward the lower end of the ranges reported here.
• Avoided transmission build-out in California. The ratios tend toward the higher end of the ranges when including the economic benefit of avoided transmission build-out in California, regardless of expectations about future generator costs and future federal tax incentives.
This study does not offer recommendations about what one should assume regarding future costs, future incentives, and avoided transmission build-out. Rather, the aim is to test the extent to which the corridor constitutes a “least regrets” proposition for major infrastructure development and its long-term benefits, anticipating how some of the most crucial variables could change by 2017.
The analysis tests four renewable resource portfolios and their likely characteristics in 2017. Modeling conducted by the California Public Utilities Commission (CPUC) provides the basis for defining the portfolios. That modeling, conducted as part of the CPUC’s Long-Term Procurement Plan (LTPP) proceedings, identified a number of “net short” procurement scenarios for CPUC-jurisdictional utilities.2
This report begins with the resources selected in the commercial interest portfolio, which gave weight to projects with power purchase agreements and for which permitting applications were substantially complete.
Two portfolios would result in meeting 33% of retail sales with renewable resources:
• the 32,184 GWh/year selected in the commercial interest portfolio (“CA33%”)
• 20,184 GWh/year of the commercial interest portfolio, and 12,000 GWh/year of Wyoming wind power (“CA/WY33%”); this portfolio excludes generic projects with no specifically demonstrated commercial interest, and assumes that some future in-state projects for which developers currently have indicated commercial interest will not meet their expected in-service dates
Two other portfolios would result in meeting 35% of retail sales with renewable resources:
• the commercial interest portfolio, plus another 4,433 GWh/year of new California resources (“CA35%”)
• 24,617 GWh/year of the commercial interest portfolio, and 12,000 GWh/year of Wyoming wind power (“CA/WY35%”); this portfolio excludes generic projects with no specifically demonstrated commercial interest, and assumes that all future in-state projects for which developers currently have indicated commercial interest will meet their expected in-service dates
Figure ES-1 shows the technology breakdown of the renewable resources that change between the California and California-Wyoming portfolios. Each change case represents 12,000 GWh of annual generation, and is replaced by Wyoming wind power sufficient to generate the same amount of energy. Large-scale solar PV accounts for the largest share of the change cases, followed by geothermal and in-state wind power.
The capital cost of the renewable energy technologies required to generate 12,000 GWh of electricity in 2017 is a major uncertainty. This study uses projections from two sources to bracket a plausible range of potential future technology costs. The 2013 Interconnection-wide Transmission Plan published by the Western Electricity Coordinating Council (WECC) provides one set of costs. These estimates were vetted and approved through WECC’s Transmission Expansion Planning Policy Committee (TEPPC). Another set of cost assumptions based on the most recent market intelligence draws on extensive input from technology experts at NREL and DOE’s Office of Energy Efficiency and Renewable Energy; these tend to be lower than the TEPPC costs. Table ES-2 shows the two sets of capital cost assumptions and their resulting LCOEs.
Another uncertainty is what federal incentives might be in 2017 and later. Current law limits the production tax credit (PTC) to wind and other eligible technologies for which construction began prior to December 31, 2013. The investment tax credit (ITC) is set to fall to 10% from its current 30% for solar and other eligible technologies placed in service after December 31, 2016. Both incentives have a history of last-minute extensions by Congress, however. To accommodate uncertainty about what these incentives will actually be in 2017, the analysis includes three sensitivities: one that assumes the PTC and ITC will be phased out or reduced by 2017 consistent with current law, one that assumes both incentives will be reauthorized at their 2013 levels by 2017, and another that assumes both are phased out completely. Table ES-3 shows the LCOEs for each change case tested in this study. The CA33% and CA35% LCOEs represent an average of the technology LCOEs shown in Table ES-2, weighted by the technology shares shown in Table ES-1. Note the relatively small differences between the CA33% and CA35% change case resources.
Each portfolio’s capacity value—the contribution that it makes towards planning reserve—is sensitive to both generation mix and the underlying weather changes that occur from year to year. Wind plants sited in Wyoming have higher capacity values than those sited in California when California wind is considered alone. When the renewable mix in California comes from a blend of wind, geothermal and solar energy, however, the combined capacity value of the in-state wind, geothermal and solar exceeds that of Wyoming wind. This is because:
• the capacity factor of the Wyoming wind is much higher than California wind or solar photovoltaic (PV). Thus, less installed capacity is required to achieve the same amount of energy delivered from the California wind and solar (energy equivalence was the driver in the creation of the scenarios).
• California in-state resources primarily comprise geothermal and solar PV. Geothermal is assumed to operate at full installed capacity during critical times, while solar PV operates at a high correlation with load during maximum load hours (maximum load hours are generally critical reliability periods). By contrast, wind energy tends to have a lower correlation with demand during peak load periods.
With respect to the BCA, the effect of adding Wyoming wind to a 2017 portfolio is a negative benefit, in that it would reduce overall capacity value between 919 MW and 957 MW. That deficit is priced using the capital cost of a natural gas combustion turbine at an estimated capital cost of $800/kW.
Production Cost Modeling
Production cost modeling (PCM) examines the operational cost of the power system in the western United States on an hourly basis over the course of a test year. PCM scenarios for this study modeled operation with and without 12,000 GWh/yr of Wyoming wind power delivered via a dedicated DC transmission line. The production costs include only variable costs such as fuel costs, variable operation and maintenance costs, and startup costs, with fuel costs being the dominant driver.
Production cost modeling of the Western Interconnection shows modest changes in operating costs when replacing a fraction of California local renewable resources with Wyoming wind power delivered to California by a dedicated transmission line: a savings of 0.2% ($31 million annually) for the 33% renewable energy scenario, and 0.1% ($14 million annually) for the 35% renewable energy scenario when Wyoming wind is included. The observed small changes in operating costs come mostly from lower startup costs when Wyoming wind power is included.
Wholesale electricity prices are reduced in most areas in response to the Wyoming wind power, but tend to be higher farther from the incoming DC line in areas that have renewable resources that could be displaced by Wyoming wind.
Benefits exceed cost across all the scenarios and sensitivities tested in this analysis. The benefit/cost ratios range from 1.62 to 3.62 across all combinations of assumptions about future generator costs, future tax incentives, and avoided transmission build-out within California.
This study applies no threshold test to the resulting benefit/cost ratios, as any determination of decision criteria is beyond the scope of this study. We note nevertheless that Order 1000, promulgated by the Federal Energy Regulatory Commission in 2011, restricts transmission utilities from using a threshold greater than 1.25 in determining whether a transmission facility has sufficient net benefits to be selected for a regional transmission plan. Savings in generator costs constitute the largest component benefit of a portfolio that includes Wyoming wind, as shown in Table ES-4. Annual generator cost savings range from around $500 million to around $1 billion depending on the assumptions about tax incentives and future generator costs.
Figure ES-1 shows the resulting benefit/cost ratios based on current laws regarding the PTC and ITC in 2017 (no PTC with the ITC reduced to 10%). The range shown on the left is based on the reference case capital costs developed by TEPPC; the range on the right uses the renewable energy cost sensitivities developed by NREL and DOE technology experts.
Figure ES-2 and Figure ES-3 show the results with the incentives reauthorized at 2013 levels in 2017, and eliminated in 2017. The ratios tend to be higher without the PTC or ITC, suggesting that the corridor may provide a hedge against the risk of reduced federal tax incentives.
The BCA also tested whether developing the Wyoming-California corridor could hypothetically pay for itself over 20 years. This was done by fully amortizing project costs over 20 years rather than the 40 to 50 years that is typical for major transmission investments, and by ignoring any benefits that the transmission line may provide over its remaining 20 to 30 years of useful life.
This equalizes the cost recovery period for the transmission asset and that of the renewable energy generation resources. The 20-year benefit/cost ratios with accelerated cost recovery ranged from 1.28 (based on the renewable resource cost sensitivity, reauthorizing federal incentives to 2013 levels, and excluding the benefit of avoided transmission build-out in California) to 2.87 (based on the TEPPC reference case capital costs, no federal incentives, and including the benefit of avoided transmission build-out in California). Increasing transmission project costs by 25% reduced the ratios, but in all cases total benefits still exceeded costs. The benefit/cost ratios ranged from 1.29 to 2.89.
The economic viability of a major infrastructure project depends on the factors tested in this analysis. The benefit/cost ratios indicate economic headroom of between $2.3 billion and $9.5 billion over 50 years on a net present value basis, as shown in Table ES-4. This range of net benefits provides an economic benchmark for examining other factors that require additional and more complex modeling, and which could not be sufficiently represented in this analysis. One such additional factor is the cost of interconnection to the CAISO network from the converter station at the terminus of the HVDC line, a cost not included in this analysis. This would require a detailed, project-specific system impact study similar to what CAISO or a transmission utility would conduct. The costs indicated by such a study could be compared with the headroom associated with the primary factors tested in this analysis, providing the next step in assessing overall project economics.
Another factor is the cost of integrating Wyoming wind power and California renewables, which is contemplated for a later phase of this study. An integration analysis would have as its objective the optimization of several decision variables, such as where to site flexible resources, possible interconnection with the AC network in Wyoming, and participation in an energy imbalance market. None of these variables could be quantified sufficiently without an integration study, but the economic headroom indicated by this analysis can help evaluate the reasonableness of different integration options. Other factors were excluded from the BCA because of time and resource limitations. These are discussed at the end of Section 1…
Summary and Future Work
This study compares the relative economics of two options for providing 12,000 GWh/year of renewable energy to electricity customers in California: a mix of California renewable resources likely to be available between 2017 and 2020, and Wyoming wind power. Either option would add to the California resources already serving customers in the state. In the second scenario, Wyoming wind power is delivered to the California marketplace via an HVDC transmission line.
Both options have discretely measurable differences in transmission costs, capital costs (due to the enabling of different generation portfolios), capacity values, and production costs. The BCA used to examine the economic difference between the two options suggest that the benefits of Wyoming wind could exceed the cost of the transmission required to deliver it under the array of future conditions tested in this analysis. Moreover, this conclusion remains robust over all future scenarios for generator cost and federal incentives that were tested in the analysis.
It also remains robust to accelerated (20-year) cost recovery for transmission, and for transmission costs 25% above the reference case assumptions. The main scenarios tested in the BCA suggest economic headroom—i.e., benefits in excess of costs—amounting to between $2.3 billion and $9.5 billion over 50 years on a net present value basis. This degree of headroom warrants further examination of costs that could not be included in this analysis. For example, the network upgrades and other system impact costs involved with connecting the southern terminus of an HVDC line to the CAISO balancing authority require more specialized analysis of CAISO flow data. The headroom indicated by the BCA provides a benchmark for evaluating the magnitude of such costs once they have been determined.
Similarly, the amount of headroom shown in this analysis provides a benchmark for measuring the integration costs that would be associated with Wyoming wind power. Because of the many options, integration issues have been set aside for a subsequent analysis.
The difference in generator costs—i.e., the capital investment required to generate 12,000 GWh/year—makes up the largest share of overall benefits. Large-scale solar PV and geothermal power are expected to be the two primary renewable resources available for new development in California after 2017, but on a dollar/MWh basis they are both more expensive than Wyoming wind. In addition, Wyoming wind power generally has a higher capacity factor than does California wind power, resulting in more energy per dollar of capital investment.
The analysis contained in this report supplements previous studies, providing a more detailed look at the transmission costs to interconnect and deliver Wyoming wind to the California market as part of a California RPS or a post-RPS scenario. Future work efforts could look at the benefits of geographic diversity and the impact on reliability. Impacts on the distribution system could also inform the cost analysis and highlight any potential reliability impacts. An analysis of the potential integration issues associated with the proposed scenarios could also further inform system costs and could include sub-hourly analyses to understand the power system flexibility requirements, potential impacts of load, solar and wind forecast errors, and changes to ancillary service requirements due to variability and uncertainty inherent in the wind and solar generation.