Much has been made about the potential of improving grid reliability and supporting a new era of distributed energy resource (DER) utilization via new pricing and grid integration business models. Locational pricing has been one of the pillars of this dialogue.
Whether in California’s distributed energy resource providers (DERPs) program, which is California PUC-approved but awaiting the implementation of final rules, or under New York’s ongoing, broad electricity restructuring process known as Reforming the Energy Vision (REV), policy makers and developers of DERs alike have pushed for access to locational pricing signals. Similar locational pricing opportunities for DERs have been brewing in Texas, currently under review by Texas grid operator ERCOT’s “Distributed Resource Energy and Ancillary Markets” (DREAM) Task Force.
With CA’s DERPs, NY’s REV, and TX’s DREAM, the U.S. may be entering a new era in which DERs—traditionally located on the retail side of the distribution grid—increasingly directly participate in wholesale power markets.
But how much value is there really in locational pricing specificity for distributed energy providers? And is it enough to turn a modest distributed energy market into a hotbed of locational pricing-targeted development? We ran some basic calculations in ERCOT to begin to find out.
Understanding locational pricing through congestion
Locational pricing differences in the transmission grid arise because of grid “congestion.” Without congestion, the entire ISO/RTO receives the same power price. That grid-wide price is determined by the intersection of the lowest clearing bid to buy energy and highest offer to sell energy. In reality, an otherwise uniform grid-wide price is complicated by congestion. Congestion occurs when a local area is effectively “walled off” from the larger grid due to grid constraints (e.g., risk of thermal overloading of transmission lines or transmission lines or substations under maintenance). The local area, in the scenario where wholesale power markets exist (i.e., within ISOs/RTOs), is then subject to a much more localized offer/bid intersection.
Congestion can push local pricing up or down. With positive congestion, local power prices are higher than in the broader grid. Power generating systems located in positively congested areas can be considered to be on the “right” side of congestion, since they help reduce power prices in the positively congested zone.
Negative congestion is the opposite, where an area of low-priced market-clearing power is walled off from the rest of the grid due to grid constraints, thus making the local pricing lower than the grid mean. Power generating systems located in negatively congested areas can be considered to be on the “wrong” side of congestion, as they are likely producing power where extra power is not needed, pushing down local prices further and potentially being forced by the ISO/RTO into outright curtailment.
What does congestion look like in practice?
Two recent screenshots taken at random moments on recent subsequent days (1/27/16 and 1/28/16) of ERCOT’s locational pricing status are provided below.
On the left, while the majority of ERCOT was experiencing pricing below $124/MWh, the very southern tip, in the vicinity of Brownsville, TX, was receiving pricing above $3,500/MWh, This was likely due to that area’s power having to come from very high offered power from the Silas Ray natural gas power plant due to the inability of other power to flow into the area.
On the right, while the majority of ERCOT was experiencing pricing of about $17/MWh, the center of Texas by the town of Commanche, or thereabouts, was experiencing pricing down to minus $22/MWh, likely due to the local 149 MW wind farm, Goldthwaite Wind Energy, producing so much power that its -$22/MWh offered energy could not be exported out of its local area. Wind power projects that earn per-MWh production tax credits (PTCs) often make negative-priced energy offers so long as they cover margin costs with the PTC plus the energy price.
Left = positive congestion. Right = negative congestion. Source: ERCOT.
What’s more, positive and negative congestion can occur simultaneously throughout the grid, sometimes in many different locations as well as over broad areas. Below is a screenshot of another randomly chosen moment on 2/18/16 showing several locations of negative congestion with negative pricing, broad swaths with pricing near $0/MWh, around $15/MWh, and in the mid-$20s/MWh, as well as a positively congested “hot spot” near Dallas with pricing above $57/MWh.
Simultaneous positive and negative congestion. Source: ERCOT.
How significant could the DER market be in congested areas?
We took a look at the most recent, complete-year real-time market data of ERCOT settlement point prices (SPPs) that we possessed (2014) and its relative congestion. We used the 574 SPPs in ERCOT that received pricing continuously in 2014. Though not the full “nodal” basis that ERCOT uses in modeling its grid (which is about 11,500 locational marginal prices; basically every substation), SPPs still provide a reasonably representative view of local, nodal congestion. The pricing distribution is shown in the histogram below.
The data showed that in 2014, 4.5 percent of ERCOT’s settlement point prices were substantially higher than the grid mean (29 percent higher, on average). How significant is this? If we’re answering the question, “Does ERCOT see a substantial percent of its locational prices well above its average price?”, the answer needs to honestly be “No.”
However, if the question is, “Could this be an interesting market for DERs to pursue locational value?”, (i.e., should DREAM enable it), the answer appears to be “Yes” based on the size of positive congestion that’s priced substantially above the market mean. ERCOT just released that its total energy served in 2015 was 347.5 TWh—4.5 percent of that is 15.6 TWh. That number is comparable to:
- More than 150 percent of all of the annual power generation in Hawaii (~9.9 TWh)
- More than 137 individual countries’ entire national annual generation
- ~105 percent of all utility-scale solar generation in California (~14.9 TWh) in 2015
- ~40 percent of all the solar power produced in the U.S. (~38.0 TWh) in 2015
Further considerations to capture locational pricing opportunities
Several additional preliminary questions need to be asked when looking at the ability for DERs to capture attractive locational pricing, including:
- How much will the newly installed distributed energy resources decrease positive congestion pricing?
- How transient is any particular point of congestion? In other words, is the congestion relatively repeatable, and if so, moderately predictable?
- If congestion does persist, how will public utility commissions consider new utility transmission lines versus, or in tandem with, DER propagation when considering congestion alleviation and maintaining reliability?
- What is the dynamic of congestion that’s most important to a DER technology? For instance, for distributed storage, the average spread in pricing over certain time intervals (e.g., might it experience significant positive and negative congestion?) could be more important than if a node has average negative or positive congestion.
- Can the node-to-hub or node-to-load zone congestion be long-term contracted?
While looking into each of these adds further insight into the “Does locational value pay?” question, for the Step One question of “Is there a reasonable volume of interesting locational pricing out there?”, the answer down in Texas appears to be “Yes.”
The grid to DERs: you complete me
DERs could well be the proverbial Renee Zellwegger’s Dorothy Boyd to Tom Cruise’s Jerry Maguire in the memorable 1990s movie, in that it’s not just that they have an interesting opportunity to participate fully in locational wholesale markets, but have been the critical piece that’s long been missing, as it “completes” the nodal, locational market design. Traditional generators don’t really locate themselves where the grid optimally needs them, as their massive generation typically dilutes out any positive local congestion that existed without them. With DERs we will have more of a “just enough” approach with DER generation, with DERs siting exactly where most optimal, finally allowing the nodal market to fully meet its purpose.
Dan Seif is a senior consultant at The Butler Firm and a former principal in RMI’s electricity practice. Chad Blevins is also a senior consultant with The Butler Firm and has been the chair of ERCOT’s Emerging Technologies Working Group since 2013.
Photo courtesy of mwwile via Flickr, Creative Commons license (CC BY-NC-ND 2.0).