Good Datasets for AI Model Training in the Power and Energy Domain

PESDPlogo

Official site

The official site of this competition can be accessed on IEEE DataPort and news will be dually updated dynamically on this site.

Abstract

Application and development of advanced AI models provide pivotal solutions for various power and energy system (PES) issues and challenges. The performance of AI models is heavily dependent on the availability of high-quality datasets. However, such datasets are often insufficiently documented. To address this challenge, we are launching the competition “Good Datasets for AI Model Training in the Power and Energy Domain”, which aims to identify, curate, and promote exemplary datasets that can accelerate research and development in PES. The competition is open to the public, inviting students and researchers at all levels to showcase datasets that demonstrate excellence in quality, relevance, accessibility, and impact on AI-driven solutions for power systems, renewable integration, energy efficiency and security, grid resilience, and sustainability.

Competition Purpose & Scope

This Competition encourages researchers across the community to contribute high-quality datasets that support AI model training and evaluation for challenging problems in PES. The competition does not constrain targeted questions (as long as they’re in the PES field), algorithms, or model choices. Instead, it emphasizes:

  • The importance of the addressed problem and the technical difficulty of applying AI;
  • The dataset’s quality, representativeness, reproducibility, and compliance for AI model training/learning;
  • Performance of the AI models trained on the dataset(s).

Submission Requirements

  • Expressions of Interest (EoI) are open until December 15, 2025. A bonus score of 0.25 will be awarded to participants who submit their EoI before the deadline.
  • All final submitted works should consist of two parts: part1 the DATASET(s) and part2 the REPORT(s)
  • All DATASET(s) must be uploaded to IEEE DataPort and comply with DataPort’s upload and publication policies.
  • REPORT(s) must include a brief introduction document and a data description article in IEEE data description journal Descriptor Articles style.
  • Please use the provided template for REPORT(s).

Find more info. in SubmissionDeatils session.

Failure to complete or submit the required materials will result in unsuccessful participation.

Key dates

Date (GMT+0)Description
Oct.15, 2025Competition launch
Oct.15, 2025Expressions of Interest (EoI)
Dec.15, 2025EoI deadline (Bonus score for participants submitting EoI before the deadline)
Dec.25, 2025Submission opens
Feb.28, 2026Submission deadline
Mar.14, 2026Administrative/compliance screening
Mar.15~May.14, 2026Review period
June, 2026 (TBD)Online defense
June, 2026 (TBD)Result announcement

Inquiries and Questions

Contact us: dppesdatacompetition@gmail.com