Knowing what happened in the past is important when preparing for upcoming similar trading periods. At Genscape, our team consistently monitors the market to establish a deeper knowledge of key market drivers and their impacts to the grid. Through lookback studies our team evaluates historical transmission outages along with the historical actual demand and supply to create a correlated point of view on how these factors impacted pricing at that time.
As you will see in the summary below, there were some considerable congestion opportunities in April of 2015. To prepare for April 2016, it is worthwhile to understand these congestion risks from last year and how they might impact trading opportunities for 2016. Typically, April demand falls into the 38-43 GW range with a peak weekday average demand of around 40 GW. Additionally, the supply stack should remain similar to March 2016 with coal and combined cycle generation comprising most of the stack and a range of 18-28 GW of generation on outage. However, knowing these two components only gets you part of the way there. To get Genscape’s point of view on April 2016 you can request a trial of our ERCOT CRR Opportunities & Insights Service which forecasts likely congestion. Additionally, consider trialing the Nodal Market Insights Application to construct your own forecasts of congestion for April and beyond.
The report below analyzes last year’s outcomes in 2015 and provides insight on how this year might be the same or different.
DAM PEAK WEEKDAY OVERVIEW
DAM Highest Priced Constraints
The table below shows the CONSTRAINTS that posted with the Top 10 highest shadow prices.
In April 2015, four of the top 10 DAM constraints by shadow price posted in the Far West weather zone with an upside impact on the Notrees Wind farm node (NWF_NWF1).
The NWF_NWF1 Node cleared $46.02 for PeakWD due to numerous line outages in the Far West weather zone including the ANDNR-BAKTP 138 KV line outage which was out from April 12 - April 19, 2015 (see Figure 1). This outage drove congestion on constraint 6596__F (HLTSW-EMATP 69 kv) during this time and elevated the Notrees wind farm consistently above other nodes in the area. The Notrees wind farm also had other dates with strong pricing including April 7 and April 9 when the 345 kv line from W_GT_345 to ODEHV was forecast to be on outage for both days. The Odessa Ector node OECCS_1 cleared $25.62 during this time for the on peak average in the DAM.
Most Frequent DAM Constraints
The table below shows the constraints which posted most often in DAM in April 2015 along with the average shadow price for all intervals it posted.
Constraint 6090_D (HLTSW-AMMFT 138 kv) posted often at 209 hours, which was also the seventh highest priced constraint during the on peak period. As you can see in the chart below (Figure 2.) the green line represents 6090_D constraint as compared with constraints 6596_F and 6915_A. All of these constraints brought upside to NWF_NWF1 throughout the month, but specifically 6090__D contributed most.
While not a constraint that posted often throughout the month of April 2015, the FAYETT_6AT2 constraint, (Fayette 345 to 138 kv) which posted in the DAM on April 20-21, had a significant impact on the FPPYD_FPP_G1 node during the on peak hours. FPPYD_FPP_G1 cleared at negative prices in the DAM over the morning and afternoon hours of both days. There was a planned outage on the FPPYD1-FPPYD2 345 kv line beginning on April 20 and scheduled through April 24, although this line outage was pulled from the scheduler and never went into outage. However, it was still responsible for driving down Fayette units 1 and 2 in the DAM with such an impact that it helped drive the average DAM price for the month considerably lower than the rest of the grid at 22.86 for PeakWD for the FPPYD_FPP_G1 node. Since the line outage was rescheduled and was never out of service, the congestion did not post in the RT.
REAL TIME PEAK WD SUMMARY
Highest Priced RT Constraints
The table below represents the constraints that posted with the top 10 highest shadow prices in RT.
Again, Notrees Wind Farm received upside in Real Time to these congestion events however it was not as significant as in DAM. Of the four constraints (6545_A, LMESA_FMR1,6090_D, ODEHV_MR1L see Figure 3.) that drove NWF_NWF1 throughout RT, only LMESA_FMR1 was in the top 10 highest priced constraints. In the RT, the price was only driven up to $30.95 for PeakWD. There were brief spikes April 7, April 17, and April 21, driving this node above the West Hub and North Hub due to these constraints but they were short lived and had a less impact on the RT than the DAM (see Figure 4).
Another node which saw the most upside in the RT was the AUSTINPL_ALL node which cleared $40 in the RT for the PeakWD. The upside was mainly driven on two days (April 8 and April 13) and due to two constraints: CKT_979_1 constraint (MAGPLANT-NORTHLAN 138 KV) and 211T147_1 constraint (GILLCR-MCNEIL 138 kv).
The strongest RT constraint by shadow price was the CKT_979_1 constraint (MAGPLANT-NORTHLAN 138 KV) which posted on April 8 with a 3500 shadow price. This helped drive the AUSTINPL_ALL node in the center of Austin to clear $2,000 for an hour on that day. This constraint was driven by a few line outages (MCNEIL-MARSFO 138 Kv, LAGOVI-MARSFO 138Kv, SEAHOLM-WARREN 138 Kv) along with strong Austin demand, and a lack of generation online (0 output at AUSTINPL_ALL node and Decker Creek steam units both offline).
On April 13 (See Figure 5 and Figure 6), similar outages (MCNEIL-MARSFO 138 Kv, LAGOVI-MARSFO 138Kv, CAMERON-FISKVILL 138 KV, NORTHLAN-WARREN 138 Kv) occurred and demand and generation circumstances drove the 211T147_1 constraint (GILLCR-MCNEIL 138 kv) which had a similar impact with upside to AUSTINPL_ALL node and LZ_AEN.
Most Frequent RT Constraints
The following table shows the constraints that posted most often in RT. The top constraint in RT was a wind driven constraint, with mainly a bearish impact to the Windthorst Wind Farm. The third most frequent RT constraint, ZO_AJO, was actually posted incorrectly by ERCOT earlier in the month, and can be seen down further in the list as AJO_ZO before they made the correction. In reality, the frequency of this ZO_AJO interface for April 2015 is 95 instances, after we add both instances together.
CONCLUSION
Bringing forth historical analysis into future planning is a key part to being prepared to react to similar or changing conditions in the tradable operating periods. Equally or more important is forecasting the upcoming congestion risk patterns. Using tools such as Genscape’s Nodal Market Insights Platform, users have a unique ability to efficiently assess historical patterns while also comparing that with forecasted views on congestion to get a jump start on future trading/hedging decisions.
The four DAM West constraints mentioned above have not posted recently. Most notably when preparing for the April Auction in early March, the system noted that these constraints did not post in the DAM in February, and only forecasted small flow increases for April of 2-4% as compared with the February results. While the positive flow increases indicate there is a better chance of seeing these constraints post this April vs February 2016, the small values of these increases mean targeting these constraints in the April Auction might only be worth the risk with a low bid.
The other three constraintsthat impacted the DAM in April 2015 were FAYETT_6AT2, 211T147_1, and CKT_979_1, only posted with minimal shadow prices in February, seen below.
After running the April 2016 CRR model in Genscape’s Nodal Market Insights Platform and comparing it to February, the PeakWD flow delta’s on these constraints are as follows:
This shows that the constraints in the Austin area are decreasing in risk, while flows on the FAYETT__6AT2 constraint are increasing 12% month over month from February. This indicates that buying a path in the CRR market targeting this constraint will likely be a better play month over month.
Using the Nodal Market Insights Platform, there are hundreds of constraints that can be studied in both the CRR auctions and in the Daily PTP markets. While these represent just a small sampling of constraints, users can study any constraints to view proprietary flow data, optimal source/sink pairs based on shift factors, historical financials as well as many other bid metrics including Genscape’s proprietary scoring metric making it easier to qualify the congestion and financial risk on paths. Click here to learn more or request a call.
The ERCOT CRR Opportunities & Insights Service is the only CRR product aimed at helping clients identify trading opportunities by focusing on the study of binding constraints, the projected flows and shift factors across those constraints, and the factors driving those flows. Leveraging this tool, clients can quickly find otherwise unknown target trading opportunities, develop new trading strategies, mitigate risk, and build new insights for the ERCOT monthly CRR Auction. Click here to learn more about the service, get more detailed performance metrics, or request a free sample report.