Making Progress Toward NY REV Goals by Adopting Probabilistic Methods for T&D Forecasting and Planning

Aug 10, 2016

The New York investor-owned utilities recently filed their inaugural Distribution Service Implementation Plans on June 30, 2016 as part of NY-REV. But the work is not done. The focus is now turning to modernizing distribution forecasting and planning for the November 1, 2016 supplemental filing.

The growth of distributed energy resources (DERs) is fundamentally changing the nature of transmission and distribution forecasting, planning, and operations. The NY Public Service Commission recognized the need to modernize forecasting and planning and has been explicit in its position that “forecasts should follow a stochastic or probabilistic methodology rather than a deterministic one.” Although utilities are not required to propose what they plan to do to implement this type of analysis until November 1, Nexant’s study for Central Hudson has already implemented the requested stochastic methodology.

Well ahead of the curve

As part Central Hudson’s DSIP filing for NY-REV, Nexant quantified the potential to avoid or defer T&D infrastructure upgrades with extremely granular data and locational insight. To our knowledge, no other utility to date has implemented a location-specific avoided T&D cost study that relies on probabilistic analysis and quantifies the option value of reducing peak demand.

One vital role of the electric utility is to ensure that electricity supply remains reliable. The New York Public Service Commission’s REV proceeding focused on unlocking the locational value of DERs – namely their potential to defer or eliminate the need for investments in traditional T&D infrastructure reinforcements. This requires quantifying the potential to avoid or defer infrastructure upgrades as granularly as possible.

Location-specific forecasting and planning methods have direct implications for DER integration.

Forecasting location-specific loads and DERs using probabilistic methods is becoming increasingly critical for T&D planning. However, local demand growth trajectories based on historical growth are inherently uncertain; those forecasts grow more uncertain further into the future. Locational, granular forecasts are also essential to establishing a more specific value of DERs and identifying locations where DERs are beneficial.

Highlights from the study

Nexant’s study of Central Hudson focuses on substation and transmission costs (it does not include circuit feeders) and was designed to meet the following objectives:

  • Analyze load patterns, excess capacity, load growth rates, and the magnitude of expected infrastructure investments at a local level
  • Develop location-specific forecasts of growth with uncertainty
  • Quantify the probability of any need for infrastructure upgrades at specific locations
  • Calculate local avoided T&D costs by year and location using probabilistic methods
  • Identify beneficial locations for DERs

The study is unique among potential studies in that it

  1. Produces T&D avoided costs estimates at a local level
  2. Uses a bottom-up approach to quantify historical year-to- year growth patterns and the amount of variability in growth
  3. Develops load growth forecasts and avoided cost estimates using probabilistic methods rather than straight-line forecasts

As a general rule, only growth-related T&D investments that are shared across multiple customers can be avoided by DERs or demand management. As loads grow, the excess distribution capacity that provides reliability dwindles. If a customer helps reduce coincident demand, either by injecting power within the distribution grid or by reducing demand, the unused capacity can accommodate another customer’s load growth, thereby helping avoid or defer investments required to meet load growth. Avoided or deferred T&D investments free up capital for other alternate uses, improving the efficient use of resources.

Not all distribution investments are driven by local, coincident peak loads. Some investments are tied to customer interconnection costs and are essentially fixed. Other investments must take place because of aging or failed equipment or the need to improve reliability and modernize the grid. These investments typically cannot be avoided by managing loads with DERs. The value of distribution deferral varies significantly across local distribution areas because of:

  • Load growth rates and anticipated changes in load curve shapes, which affect whether infrastructure upgrades can be avoided and how long they can be deferred
  • The amount of existing excess capacity or the amount of additional load that can be supported without upgrades
  • The magnitude, timing, and cost of projected distribution upgrades
  • The design of the distribution system

In areas with excess distribution capacity—or areas where local, coincident peaks are declining or growing slowly—the value of distribution capacity relief can be minimal. In areas where a large, growth-related investment is imminent, the value of distribution capacity relief can be quite substantial, especially if it is possible to delay or defer distribution infrastructure upgrades for a substantial time. However, many Central Hudson distribution areas have declining or slowly growing loads, or they have sufficient capacity already built such that distribution investments are not needed in the foreseeable future.

The key findings from the analysis are:

  • Most substations and transmission areas are experiencing declining loads or have ample room for growth over the next 10 years.
  • The expected avoided costs vary by location and year and are highly concentrated. Avoided costs are realized if additional resources are placed in the right locations. Without targeting, the value of distributed resources is diluted.
  • For many distribution substations and transmission areas that have expected growth, the potential for avoided infrastructure upgrades through DER resources is minimal because there is already sufficient capacity built in the area to meet load growth.
  • The avoided cost estimates reflect the uncertainty in the forecasts and the risk mitigation value of demand management.
  • Despite a low likelihood of exceeding design rating in the next 10 years, DER resources can provide risk mitigation value at targeted transmission areas and substations if they are at the right locations, target the right hours, and are available at the right times.
  • In practice, all avoided T&D costs are location specific. For system-wide untargeted values, the estimates take into account the likelihood reductions would be in locations with value due to random chance. Without precise targeting, the likelihood that reductions defer or delay transmission upgrades is relatively low.

The study demonstrates the value of developing T&D avoided cost estimates at a local level using probabilistic methods. Because the methodology is relatively novel, it may require future refinements and improvements. Future studies can be further bolstered by conducting sensitivity analyses and refinement of  engineering rules, which trigger T&D infrastructure upgrades.

Come hear me speak on this topic

I’ll be presenting this material at the Southwest Forecasting and Customer Analytics conference on September 15-16, 2016 at Tucson Electric Power headquarters.