Key Information

Register
Submit
The challenge is finished.

Challenge Overview

Prize Distribution

1st place - $2,000

2nd place - $1,200

3rd place - $600

4th place - $300
 

Checkpoint Prizes: $100 each for up to 5 winners. The checkpoint submissions due at Sep 23, 12:00 PM EST

 
 

Challenge Overview

We are looking for solutions that can compute Yaw Misalignment angle of Wind Turbines from the Supervisory Control and Data Acquisition (SCADA) data, which can be later validated by the Yaw misalignment values from Light Detection and Ranging (LiDAR) data. Ideas, insights, datasets, and suggestions are wanted! We hope to get great solutions from you, Good Luck!
 
Please note that this is a data science based ideation and PoC challenge - there will be a checkpoint you can submit for in order to receive feedback and potentially one of five $100 prizes - we've detailed exactly what we're expecting as outputs below.

Background

The business objective is to optimize energy generation at each WTG by dynamic calculation of Yaw misalignment. We need a predictive model for dynamic calculation of Yaw misalignment at every 10 minutes rolling for next 7 days by using historical SCADA Data and calculate optimum value for the period.
 
In wind turbines, correct nacelle alignment to main wind direction is necessary for optimum generation and improve the annual energy production. Lot of analysis & reports says that 4 deg average will cause 1 % AEP loss
 
Ideally, Yaw offset should be zero degrees, but in a conservative way, business would like to keep the band within 0 to 2 degrees on either side.. The ultimate goal of this project is to optimize energy generation at each Wind Turbine Generator (WTG) by dynamic calculation of Yaw misalignment at every 10 min interval.
 

At WTG Unit level

  • Using 10 min aggregated historical SCADA data Yaw misalignment angle A (as shown in Fig 1) needs to be calculated for the next 7 days.


Fig 1: Yaw misalignment

 

    (B+A) tends toward C

Where

    A = Yaw misalignment angle

    B = Wind direction measured by wind vane recorded by SCADA (Relative Wind Direction)

    C = True Wind Direction wrt Nacelle Position

 
  • Effect of the wake to be incorporated. Since the turbines extract the energy from the wind, the wind leaving the turbine blades have lower kinetic energy compared to free wind entering the blades. The model should consider the effect of wake in Yaw misalignment calculation.

 

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. The improvement in generated power through yaw correction based on the dynamic Yaw misalignment values will be measured through the power curve.


Power curve will be the means to measure the effectiveness of Yaw misalignment correction. It indicates the relationship between wind speed and Actual Power generated. If the wind vector is perpendicular to the rotor area, the turbine yields the optimum performance and the large inflow angles to the rotor lead to lower the performance i.e. higher Yaw misalignment. Below image illustrates through a typically measured power curve, before Yaw misalignment calculation, and after Yaw misalignment calculation

  In this ideation challenge, let’s focus on one wind farm -- The wind farm lies on two parallel hill ridges aligned in the northwest to south-west directions. As measured with GPS the surrounding base height and maximum elevation of the hill ridges are 720 m.a.s.l (Meters above sea level) and 860 m.a.s.l respectively. The ridges are with steep side slopes descending to almost flat open plains of surrounding area.
 

Key Data challenge:
  • We have SCADA data spans over 5 years, while the LiDAR data is collected only for 1 turbine for a duration of 2 months ; 
  • We have an option of installing LIDAR on the turbine cluster, however we are not in favor of using LIDAR as a solution to this problem. Hence we would like to use SCADA data to build a model that can be leveraged for all turbines to predict the YAW misalignment for each of the turbines separately.

Task Detail

  In this Ideation challenge, we are looking for ideas for the following problems:
  1. Given the SCADA data, how can you compute the Yaw misalignment angle shown in Fig 1 for next 7 days? What if we want to run this as a Marathon competition?

  2. If the model is created using the existing SCADA data, what will be the ground truth and how to create the same?

  3. There exists a Wake effect ( Unit level ): the extraction of energy from wind across a turbine rotor produces an aerodynamic wake region downstream from the rotor. How to let the computational model implicitly take the Wake effect into the consideration? So far, we don’t have data about this effect.

 

There are some related papers, for example:

  • Mittermeier, Niko, and Martin Kühn. "Determination of optimal wind turbine alignment into the wind and detection of alignment changes with SCADA data." Wind Energy Science3.1 (2018): 395-408.

 

Also, we are open to leverage those public datasets. You may find the recently released Google Dataset Search useful.

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

Data Description

  • The SCADA data shared contains abundant information over a year. There is a separate spreadsheet discussing the meaning of each column.
  •  LIDAR data for 1 month mapped to same timeline of SCADA data
 


Final Submission Guidelines

Submission

Commercialization 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 following contents. The client will judge the feasibility and the quality of your proposed likelihood function.

  1. Title : Title of your idea

  2. Abstract / Description : High level overview / statement of your idea

    • Outline of the approach of the algorithm implementation

    • Outline of each functional unit of the algorithm implementation

    • Summarized conclusion

  3. Details :

    • Detailed description. You must provide details of each function unit and details of how it should be implemented

      • Description of the entire model approach to calculate the YAW misalignment angle and mechanism
      • The advantage of your idea - why it could be better than others

      • If your idea includes some theory or known papers;

        • Reason why you chose

        • Details on how it will be used

        • Reference to the papers of the theory

      • Reasonings behind the feasibility of your idea

    • What is the Ground truth and approach to create the same, Pseudo code(or PoC code), and its description

  4. Appendix(optional) :

    • Bibliography, A reference to the paper, etc.

 

Checkpoint Submission

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

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 compute and predict the Yaw misalignment angle (single value for the complete duration of data) from the SCADA data which will be validated against Yaw misalignment angle from LIDAR data ?

    • 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?
    • Does your solution consider the Wake effect?
  • 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.

 

Note: We will be looking mainly at “Abstract / Description” for the checkpoint submissions.

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.

Supplementary materials

You will be able to download supplementary materials after registering to this challenge.


 

ELIGIBLE EVENTS:

Topcoder Open 2019

REVIEW STYLE:

Final Review:

Community Review Board

Approval:

User Sign-Off

SHARE:

ID: 30071178