Titan Eye - Energy Forecast Wind Ideation Challenge

Challenge Details

We are looking for solutions that can forecast energy for Wind Energy operations based on the Supervisory Control and Data Acquisition (SCADA) data, power curve information, and weather forecast data (wind direction, wind speed, pressure, temperature, humidity, air density, etc) for multiple sites at a turbine level. We hope to get great solutions from you, Good Luck!

Project Background

We are currently solving multiple data science use cases with Titan Eye. This use case aims to forecast energy generation for Wind Energy operations.

The ability to forecast energy generation accurately is critical to Wind Energy operations for achieving energy delivery commitments and statutory compliance. Wind energy generation forecasts are currently being procured from 3rd party provider. The business objective from this use case is to:

• Better Maintenance Scheduling.
• Scheduling and forecasting of Energy Generated.
• Budgetary Purpose for management.

Accurate Wind Energy Forecasting helps in:

• Reducing imbalance charges and penalties due to statutory compliance. 
• More efficient business operations, and maintenance scheduling at site and turbine level. 
• To make efficient budget decisions.

The task here is to build most accurate hybrid weather forecast model for various times of the year for each of the three wind sites by using the weather data from identified 3rd party weather forecasters to forecast energy production using WTG power curves. 

This requires analyzing historical trends/patterns in weather parameters like wind speed, wind direction, pressure, temperature, humidity etc. and then establishing a correlation with designed power curve to arrive at most accurate energy forecast for each turbine.

Challenge Details

In this ideation challenge, we are looking for ideas for the following problems:

• Suggest the models that can be used for energy forecasting including the details on how to measure the accuracy of the model after thoroughly analyzing the provided SCADA and WTG Power Curve information. Also the sample energy forecast from the existing vendor is shared in the forums to get an idea of the expected output.
• Suggest other features which are required for accurate energy forecast
• Identify the most reliable 3rd party (public domain) weather forecasts for each site and time of year. Also some pointers on how to calculate the accuracy of the weather forecast is needed.
• We need the weather forecast specific to the turbine location, so you have to suggest the best temporal and spatial resolution of data that need to be used for our model.
• Is weather forecast required at turbine hub height

The final deliverables is a report explaining the algorithm, sample ground truth, pseudocode or PoC that will demonstrate your idea. More details are discussed in Final Submission section.

Power Curve

The power curve of wind turbine indicates the relationship between output power generated by the wind turbine and different wind speeds at hub height. Power curve helps in energy assessment and performance monitoring of wind turbines. This graph is plotted between normalized wind speed and average power generated at a turbine / site level. This information can be used to compare the weather forecast and the average power output for a time interval.



Here is a very simple forecasting example, I know my wind speed forecast for tomorrow is  8 m/s, I will check data from the site specific power curve. I get how much power is produced for 8 m/s. The power corresponding to 8 m/s is  my power forecast for tomorrow.  We need to take care of revisions as well in case if my weather data has changed. Most APIs weather Data is revised on frequent intervals.

Data Description

• The SCADA data shared contains abundant information over a year. There is a separate spreadsheet discussing the meaning of each column.
• Sample Energy Forecast data that can be referred.
• Power Curve Information is shared

The data can be downloaded from the forums after registering for the challenge.

Important Notes

• At Daily Level, we need to forecast energy at 15 minutes interval at 10 PM on day ahead basis for each day. There will be 96 blocks for daily energy forecast for each turbine.
  • Daily: 00:00 -23:45
  • Frequency: 96 block per day at 10 PM on day ahead. 
• At Weekly Level, we need to forecast energy at 15 min interval for each day for next 7 days. There will be 96*7 blocks for weekly forecast.
  • Daily: Day 1 – Day 7
  • Frequency: 96 block per day – Generated every day at 10 AM
• At Monthly Level, we need to forecast energy on daily basis for each turbine for next 60 days.
  • Monthly: Day 1 – Day 60
  • Frequency: 1 block per day for – Generated at 10 AM 1st of Every Month
• At Annual Level, we need to forecast energy on monthly basis for each turbine for next 12 months.
  • Annually: Month 1- Month 12
  • Frequency: 1 block per month rolling upto next 12 months

 

Final Submission Guidelines

Commercial Requirements

Note that an important goal of this project is to build a commercial software. If you want to use some open-source softwares that can not be used for commercial purposes, please also discuss the effect in your submission and show the development potential of alternative code.

Contents

A document with details for the proposed algorithm and/or a proof of concept solution, pseudo-code or any documentation/ previous research papers that helps illustrate proposal.
 
The final submission should be a report, more like a technical paper. It should include, but not limited to, the ideation report template. The client will judge the feasibility and the quality of your proposed likelihood function.

Checkpoint Submission

In this challenge, we allow checkpoint submissions. In the checkpoint submission, please at least include the “Abstract / Description” part. 

Check point Deadline: 20 Feb 2019, 09:00 EST
Checkpoint Prize: $100 for 5 submissions

Final Submission

In the final submission, you must submit all items described in Contents section above.

Format

• A document should be minimum of 2 pages in PDF / Word format to describe your ideas.
• It should be written in English.
• Leveraging charts, diagrams, and tables to explain your ideas is encouraged from a comprehensive perspective.

Judging Criteria

You will be judged on the quality of your ideas, the quality of your description of the ideas, and how much benefit it can provide to the client. The winner will be chosen by the most logical and convincing reasoning as to how and why the idea presented will meet the objective. Note that, this contest will be judged subjectively by the client and Topcoder. However, the judging criteria will largely be the basis for the judgement.

• Feasibility and Completeness of the Idea (60%)
  • Can your solution forecast the energy accurately from SCADA, power curve info and weather forecasts
  • Approach for creation of ground truth
  • Considering we may launch a Marathon challenge in few weeks; Does your solution cover how to run a Marathon challenge using the current data or adding more public datasets?
• Ease of implementation (40%)
  • That pseudo code or PoC code is presented
  • The above code Implementation is easier than other submissions

For example, following will be judged higher

Having PoC code is better than just pseudo code, but if the pseudocode report has better solution and highly feasible reasonings to accurately calculate the scores, the latter report will be evaluated better.

Submission Guideline

You can submit at most TWO solutions but we encourage you to include your great solution and details as much as possible in a single submission.