1. The case
The Electric Reliability Council of Texas (ERCOT) manages the flow of electric power to more than 26 million Texas customers — representing about 90 percent of the state’s electric load. As the independent system operator for the region, ERCOT schedules power on an electric grid that connects more than 46,500 miles of transmission lines and 650+ generation units. ERCOT organizes prices by:
- a) Load Zones (LZ): A group of Electrical Buses. Every Electrical Bus in ERCOT with a Load must be assigned to a Load Zone for Settlement purposes. LZ_AEN, LZ_CPS, LZ_HOUSTON, LZ_LCRA, LZ_NORTH, LZ_RAYBN, LZ_SOUTH and LZ_WEST.
- b) Trading Hubs (HB): An energized Electrical Bus or group of energized Electrical Buses defined as a single element. The Locational Marginal Price (LMP) of the Hub Bus is the simple average of the LMPs assigned to each energized Electrical Bus on the Hub Bus: HB_BUSAVG, HB_HOUSTON, HB_HUBAVG, HB_NORTH, HB_PAN, HB_SOUTH and HB_WEST. The Locational Marginal Price for each Settlement Point, normally produced by SCED (Security Constrained Economic Dispatch) every five minutes.
incorporating the cost of producing and delivering energy from individual nodes, or locations on the grid, where the transmission lines and generation interconnect. ERCOT also publishes the DAM System Lambda, Day Ahead Market System Marginal Price which is LMPs minus system loss and system congestion at a given node.
ERCOT allows trading at trading hubs, in an intraday 5-minute market (SCED), as well as an hourly DA market.
This case study reviews a day-ahead (DA) price forecast model for HB Houston LMP price, at an hourly resolution. The price forecast model is implemented using QR AI Forecaster.
The DA market opens for bids and schedules seven days before and closes the day prior to the trade date at around noon. We therefore execute and publish our DA forecast at 8:00 a.m. market time for the next day, to allow ample time for traders to plan their DA trading strategies. QR AI Forecaster serves as the key input of QR Trading Optimizer to optimize the bidding of your assets.
The data for this case study comes from QuantRisk free and publicly available forecast webportal. You can access the performance of our price forecast models for multiple markets here.
2. The Data
The hourly time series of HB Houston LMP prices is the main data we use to train the AI model. Feature engineering is key to increasing the accuracy and stability of price forecasts. For these we use various demand and supply side data published by ERCOT.
Input Data. The AI forecast model in this case study uses 4 months of hourly data for training, from March 25th to July 25th:
- Actual data: HB Houston LMP prices data is published by ERCOT between 12:00 and 13:00 for the next day. This is the price we are going to forecast.
- Demand side data: this is our own DA System demand forecast for HB Houston.
- Supply side data: Different DA solar and wind generation forecasts, and total outages (TotalResource, TotalIRR and etc.) data all published by ERCOT.
Output data. The output of this AI forecast model is hourly DA price forecast for HB Houston executed and published at 8:00 a.m. Chicago time for the next day.
3. Data Exploration
The first step in designing the AI model is to take a deeper look into the data and identify quantifiable behavioral patterns that may not be visible at first glance.
In order to do that, the QR data science team uses different data analysis and visualization tools to extract information from raw data. You can see below the 4 months hourly HB Houston DA price used for training.
The above price data shows clear seasonality change in the value of the peaks. These were due to weather factors and the heat wave leading to increased demand and as a result prices. We need more granular analysis to define the right features for this model.
The plot above displays the hourly HB Houston price averages across the 4 months training data ranging. This data can help us see the daily pattern and daily Peak and off Peak.
Conclusions:
- The HB Houston price doubled peak is in the evenings at 4:00 p.m. and 7:00 p.m.The forecast model is required to capture these.
- This plot suggests to split the day in 3 periods and train the model on each data set, midnight to 12 p.m., 12 p.m. to 8 p.m., and 8 p.m. to midnight. This can be automatically accomplished by checking the Model Splitter feature of QR AI Forecaster.
- The price is clearly hour dependent. We therefore must use the built-in feature Hour-of-the-day.
We can further refine the above analysis by drilling down to days of the week. The next plot displays the average hourly price for each day of the week, computed for the 4-month training data.
- a) The pattern of data is almost the same in terms of hours of the peaks in different days of the week. However,
- b) Almost all the weekdays have different levels of price data in the peak area. With Mondays having the highest peak (even considering the small peak at around 6:00 a.m. and 7:00 a.m.) followed by Sundays.
- Weekday as a built-in feature or predictor to the AI Model to indicate the days of the week.
- Also Hour and Is Working Day built-ins to reflect other patterns within a day.
4. Feature & Correlation Analysis
To select the best external time series as features or predictors, we need to analyze their correlation with the main price time series HB Houston we are forecasting. A good or useful feature must be closely correlated to and shed light on a particular behavior of the main time series.
The ERCOT DA price data, including HB Houston, present several challenges in the 4 months period considered in this Case Study. Our data science team resolves these by configuring the right AI model so as to produce accurate price forecasts for our clients.
Factoring in demand effects
As an example considering the current case study, in the image below you can see that the prices have higher peaks starting from around June 15th. As expected it is correlated with the demand of the ERCOT Total that is plotted in the image below and you can find an increasing trend with higher levels in demand starting from June. The designed models should capture the trends like this which is handled by QR data science team. This was uncovered by our data science team, using our data analysis toolbox, they executed correlation analysis across many time series data published by ERCOT against the main HB Houston DA price.
Conclusion:
- The Outlier Treatment feature of QR AI Forecaster needs to be enforced for this HB Houston DA price forecast model.
- ERCOT demand time series should be included as an external feature or predictor in the AI model, to guide the forecast in predicting such irregularities.
Conclusion:
- The Outlier Treatment feature of QR AI Forecaster needs to be enforced for this THSP15 DA price forecast model.
- CAISO Solar Generation time series should be included as an external feature or predictor in the AI model, to guide the forecast in predicting such irregularities.
5. The AI Forecast Model
By now the preliminary data analysis work has been done and we are ready to implement the best AI model for this case study, considering the specific features and data discussed previously. We use QR AI Forecaster cloud service. This platform is a no-coding automated AI platform, where modeling is done by dragging and dropping various model and feature engineering components in an intuitive AI dashboard, to automatically assemble and execute the final forecast model.
1) Data Preprocessing
- a) Gap Filling can be accomplished by several methods (Linear Interpolation, Weekly Pattern, Daily Pattern, etc. ), for price data we use Linear Interpolation.
- b) Outlier Detection can be accomplished by several methods. For the current price forecast case study, we use a standard deviation method, removing data at 3 SD, and replacing it with local average.
- c) Calendar is used by a processor to swap mid–week holiday data with Sundays in the future, and with a regular day in the past during forecast and training respectively, as well as labeling working days and weekends for the AI modeling.
2) Feature Engineering
We configure the following features in QR AI Forecaster dashboard:
| Predictor Type | Features |
|---|---|
|
Lagged Predictors |
Previous Point (which here is previous hour) |
|
Built-in Predictors |
Hour, Enhanced Weekday, Week Number, Month, Is Working Day. |
|
External Time Series Predictors |
Demand forecast, Wind forecast, Solar forecast, Outage and QR designed formulas with combination of predictors |
3) AI Model
- a) Data Augmentation:
- Data augmentation in data analysis is a technique used to increase the amount of data by adding slightly modified copies of already existing data or newly created synthetic data from existing data. It acts as a regularizer and helps reduce overfitting when training a machine learning model
- We are using augmentation in this job, the data has been scaled by 10%.
- b) Model Optimization:
- We configured a range of deep learning AI models for this case study. They have their own advantages. We present here 3 machine learning models, NGBoots, LightGBM and XGBoost. These can be configured to perform with nearly equal accuracy levels.
For example with XGBoost we configure the following 12 hyper-parameters:
| booster | number of estimators | gamma |
|---|---|---|
|
max depth |
min child weight |
max delta step |
|
Subsample |
l1 regularization coefficient |
l2 regularization coefficient |
|
base score |
evaluation metric |
objective |
- c) Model Execution:
- Once the model is finalized, training one of the machine learning models over 4 months of hourly HB Houston DA price data, and executing a DA forecast takes about 1 minute.
6. Forecast and Accuracy Analysis
1) The Forecast Display Dashboard
QR AI Forecaster has several data visualization dashboards that gather in one screen:
- a) Our forecast HB Houston DA price data, computed and published at 8:00 a.m. CDT.
- b) The actual HB Houston DA price data, published between 12:00 and 01:00 p.m., after the DA market had closed around noon.
- c) Forecast error is computed by MAPE (mean absolute percentage error), and MAE (mean absolute error). These are listed in table format and gauges.
2) Accuracy Analysis
- The ERCOT DA trading hub HB Houston price forecast has a MAPE of 10.5%.
- Other ERCOT trading hubs DA price forecasts, using the same machine learning model, have a similar accuracy.
- As you can see in the other Case Studies we have published for DA price forecasting in MISO and PJM, using the same machine learning model, with different parameters and features, results in a MAPE of 6 to 7%. This is mainly due to the fact that:
- The dynamic of price level change from one day to the next can be very different across markets. E.g. CAISO DA prices are a lot more correlated to fuel.
- Each market publishes different sets of supply and renewable data. Some are better correlated to DA prices.
Download Brochure
QR AI Forecaster offers a wide range of ready-for-use, advanced AI forecasting solutions, learn more about our forecasting solutions.
