Risk Metrics and Analysis

RISK 00001
Published On: 02/23/2021

Question: Can DESC elaborate on the updated quantitative risk analysis and how it is applied to the company's preferred plan?

Answer: Please see the 2020 Modified IRP for further details.  This should be publicly available prior to the next Stakeholder Advisory Group Meeting.

RISK 00002
Published On: 02/23/2021

Question: On slide 44 of the Session I Stakeholder Advisory Group presentation, why is levelized cost a metric instead of net present value? Can you define all of the metrics on this slide, e.g. reliability?

Answer: This is an error on the slide, levelized cost should be replaced with levelized net present value. DESC defines the metrics on slide 44 of the Session I presentation as follows:

  • Levelized Net Present Value: The Levelized Net Present Value metric is a comprehensive measure of the relative costs to customers of each of the fourteen resource plans over the 40-year period from 2020-2059. The comparison is based on the forty year levelized net present value of the incremental costs of each resource plan. The incremental costs include incremental operating costs, capital costs for new generation, incremental capital costs for ongoing operation and maintenance, and DSM costs.
  • CO2 Emissions: The CO2 Emissions metric compares the expected emissions from the fourteen resource plan as forecasted at the end of 40-year period ending in 2049. 
  • Clean Energy: The Clean Energy metric compares the fourteen resources plans based on how much energy they produced as forecasted at the end of 40-year period ending in 2049.
  • Fuel Cost Resiliency: The Levelized NPV Fuel Cost of generation plans as modeled in the Modified 2020 IRP fully captures fuel costs and anticipated changes in fuel costs over a 40-year planning horizon for each plan. As a result, the Levelized NPV Fuel Cost metric provides important data about how plans perform in the face of fuel price changes. 
  • Generation Diversity: Each of the resource plans modeled assumes the addition or retirement of different suites of generation sources.  For that reason, each of the plans results in a different level of generation diversity at the close of the 40-year planning period. The generation diversity of each resource plan is ranked according to the percentage that the generation mix it creates is concentrated in any one type of generation asset.   
  • Reliability Factors: DESC has identified a set of reliability factors that measure the generation types’ ability to supply certain ancillary services, operating characteristics, and capabilities and meet certain locational considerations that support grid requirements in normal operations and in restoring power after storms or outages.
  • Mini-Max Regrets: The Mini-Max Regret analysis evaluates each resource plan against the lowest cost plan in each scenario and calculates the difference in the 40-year levelized NPV between the plans. The maximum change from the best plan in each scenario sets the max regret score for each resource plan.
  • Cost Range Analysis: The Cost Range Analysis evaluates the variation in the 40-year levelized NPV for each plan across the 27 scenarios that were modeled.  The maximum variation for each plan sets the score.

RISK 00003
Published On: 04/26/2021

Question: Although I don’t have a strong opinion on which risk metric approach is preferable, I do feel strongly that stochastic analysis is often not the best way to capture risk. I prefer a scenario analysis with a range of scenario-based outcomes. I’m not a fan of the technology risk metric. This metric comes from the need to be concerned with fuel risk, but as we move away from that, it’s less necessary. I believe the diversity of resources is a better metric.

Answer: The DESC IRP team agrees that stochastic analysis has to be properly implemented to be significant. We also agree that risk associated with some technologies are fuel related, which is a factor often considered in stochastic analysis, but some related to technology risk are often not considered. DESC’s IRP analysis uses scenarios, consistent with this observation, to consider a wide range of factors

RISK 00004
Published On: 04/26/2021

Question: Doesn't reliance on purchases also reduce the risk of being reliant on stranded assets?

Answer: We recognize that there are potential risks and benefits of reliance on purchased power. With a greater reliance on the market comes less reliance on owned assets. Therefore, if DESC owns assets that operate below the cost of the market there can be advantages. On the other hand, if DESC owns assets that are above the costs of the market, this can strand assets. 

RISK 00005
Published On: 04/26/2021

Question: Is commodity price risk specific to fuel costs only or are you considering broader commodity risk (steel as an example)?

Answer: The Commodity price risk metric is used to evaluate the cost risk associated with fuels burned; it does not include steel. DESC assumes new generator costs, including steel prices, rise based on a Handy-Whitman index when evaluating portfolios in the IRP.

RISK 00006
Published On: 06/11/2021

Question: Given recent events in Texas, are potential fuel supply interruptions part of the reliability analysis?

Answer: Yes. Our natural gas units rely on multiple pipelines from shale gas sources from the Gulf coast and several have oil fuel backup. Additionally, coal maintains a 60 to 90-day fuel supply.

RISK 00007
Published On: 06/11/2021

Question: Are you using 2049 as a 1-year snapshot on carbon emissions? Because cumulative emissions throughout the period will cause cost risks to ratepayers if CO2 is regulated.

Answer:

The impact of cumulative emissions are captured in the CO2 costs incurred by each different portfolio. DESC will consider reporting a cumulative CO2 table into the outputs. 

RISK 00008
Published On: 06/11/2021

Question: Is the CO2 metric cumulative over the entire planning period or just in the year of 2049?

Answer: The CO2 emissions metric measures the portfolio’s 2049 emissions as a measure of progress towards DESC’s 2050 target. 

RISK 00009
Published On: 06/11/2021

Question: Have you considered tracking water intensity as a core metric?

Answer: DESC does not consider the water intensity of the portfolio as a core metric but will take that suggestion into consideration. 

RISK 00010
Published On: 12/21/2021

Question: Are forced outages more likely to occur when load is high-- i.e., when more units are required to be online and are more likely to be running flat out?

Answer: The company is not aware of a study that forced outages at individual units are more likely to occur during periods of high load. We welcome any further information that Stakeholders would like to provide on this topic. DESC reports forced outages in the NERC GADS database, some of which are weather related.

RISK 00011
Published On: 03/18/2022

Question: Is the new expected Vogtle output any significant part of the transmission constraints at Jasper? I assume that new output is included as baseload.

Answer: All Vogtle units are dispatched during peak summer and winter cases and that has an impact on the Dominion system.

RISK 00012
Published On: 04/20/2022

Question: Including interchange in the reliability study, I think it's important to make the distinction between relying on your neighbors for reliability and accurately modeling the flow of power. For example, if there has historically been trading with other BAs then why wouldn't you include that in this analysis? And you could explicit model other BAs or otherwise set up constraints so that there wouldn't be interchange unless power is actually available. The concern is that NERC requirements are on a 30-min interval, whereas DESC might be over constraining the system with 5 min intervals, resulting in more unserved energy events if looking purely in isolation. We typically don’t have generation inadequacies on 5-min intervals.

Answer: If DESC was planning on using off system resources to maintain reliability, they would be included in the study. DESC found that resource adequacy reliability events tend to impact its neighbors at the same time as they impact the Company. So that when DESC has had to call on resources in the past to deal with reliability issues, it was unable to rely upon them.

DESC is currently studying the system at an hourly interval within PLEXOS.

RISK 00013
Published On: 04/20/2022

Question: Thermal forced outage rates are not one of the variables tested, they are assumed to be static?

Answer: Yes, the outage rates at thermal units are assumed to be static. DESC uses a rolling 5-year average, which is not dissimilar to the reserve margin contribution assigned to energy limited resources.

RISK 00014
Published On: 04/20/2022

Question: Did Dominion rely on imports recently when its nuclear plant had an unplanned outage?

Answer: VCSNS came offline at 5:28 p.m. on November 15, 2021 and remain offline until 7:48 p.m. on December 10, 2021. DESC did import energy for many of the days during the outage period. These purchases were due to several factors that are explained in testimony, rebuttal testimony, and responses to audits and requests in South Carolina PSC Docket No. 2022-002-E.

RISK 00015
Published On: 04/20/2022

Question: Are the reliability considerations in the retirement study different than the metrics you've used previously?

Answer: DESC may use different reliability considerations in the retirement study than it does in future IRPs. This is because the retirement study is attempting to evaluate whether reliability can be maintained in the short term, while the IRP is for long term planning.

RISK 00016
Published On: 06/17/2022

Question: It would also be useful to explore how many of the minor violations could be mitigated with other NWA, such as dynamic rating, operate around, selective reinforcements, and other grid enhancing technologies.

Answer: The TIA was a preliminary assessment indicating the contingencies shown to be the most severe for each limiting element listed. Evaluation of many more contingencies that created overloads or high loading of transmission elements would be needed for op guides, etc. This all takes time and will ultimate be addressed in future System Impact studies. Winter ratings were included as part of the winter season studies. Dynamic ratings are a short-term operational tool. Operations will use dynamic ratings as appropriate facing often conditions worse than N-1-1 contingencies.

RISK 00017
Published On: 06/17/2022

Question: We recommend that DESC provide detailed heat rate modeling assumptions for all units at the next stakeholder session. In the case of the new ICT and CC generators, the Company’s specific heat rate curve was not properly refit to the polynomial curve. To avoid this change – and use the Company’s heat rate curve directly, the model’s “Production object” setting for “Max Tranches” must be set to less than three so that the simulator used the marginal heat rate function provided in the input data verbatim.

Answer: DESC has corrected the issue and is specifying efficiency with an average heat rate curve.

RISK 00018
Published On: 06/17/2022

Question: We would like to understand if and how DESC is evaluating the variability of outages across time, correlation between outages at thermal generators, the effects of weather on outages, and fuel supply related outages on its thermal units. We ask DESC to accredit thermal generators at their seasonal rating less their forced outage rates (an approximation of unforced capacity or UCAP).

Answer: Forced outages are spread randomly across the year. These are not based on seasonal weather, since DESC has not been able to find a strong correlation between seasonal weather and outages.

Thermal resources are accredited with their summer and winter capacity. DESC disagrees that these units should be discounted and that there is a bias. Solar and battery resources have forced outage rates that are not used to discount their capacity contribution. The ELCC for solar and battery resources is based on their limited dispatchability. Thermal resource don’t share this limited dispatchability.