This competition focuses on wind turbine-level wind power forecasting.

It provides a wind power forecasting (WPF) dataset covering two wind farms (each with multiple wind turbines), split into a training set and a test set.

Using the training data, participants must predict the output power of each turbine in the test set for 288 consecutive future time steps (total 2 days).

The primary objective is to develop a predictive model that uses the limited future meteorological variables to accurately predict the future active power output of each wind turbine.

A successful model will be evaluated by a panel of judges based on:

1. The accuracy of the methods;

2. The suitability and innovation of the used methodology;

3. The insights derived from the model;

4. The clarity, technical correctness and completeness of the report and the presentation.

Eligibility Rules: 

1. Teams of up to 3 participants may participate in this challenge.

2. To qualify for the final round submission, teams must submit predictions in the first round, which allows participants to improve model performance based on round-1 feedback.

3. Important: Each finalist team must have at least one member register for the 4th ICQSR Conference and attend the data competition workshop to present live at the workshop on July 5.

Registration:

1. Registration Opens: April 15, 2026

2. Registration Deadline: May 30, 2026 (AOE)

3. Registered participants will receive the dataset download link after the registration.

All submissions should contain a single zip file named ICQSR2026_Data_Teamname.zip and emailed to Yanting Li (ytli@sjtu.edu.cn) by June 15, 2026, anywhere on earth.

The submission must include:

1. Technical Report: Should not exceed 25 pages (A4 size/letter size), excluding references.

   • It should clearly highlight your methodology, assumptions, preprocessing steps, and implementation details.

2. Source Code:  Include the source code used to generate figures and tables in the report.

   • Participants can use any coding language to solve the problem.

3. Prediction Outcome:  Please put all predicted measurement values on the test data in a single CSV file following the data format of the output data.

Yanting Li, Professor, Department of Industrial Engineering and Management, Shanghai Jiao Tong University, Shanghai, China. Email: ytli@sjtu.edu.cn

Sukjoo Bae, Professor, Department of Industrial Engineering, Hanyang University, Seoul, Korea. Email: sjbae@hanyang.ac.kr

Chen Zhang, Associate Professor, Department of Industrial Engineering, Tsinghua University, Beijing, China. Email: zhangchen01@tsinghua.edu.cn