마이크로칩 9월
This page was machine-translated and may differ from the original. View original

SolarOnCare, POSTECH, KAIST, and UNIST to Host Renewable Energy Generation Prediction Competition

Google 우선 소스Published2026.09.22 09:42


 
Total Prize Money of 17 Million Won; VPP Optimization Task Using Operational Data from 253 Solar Power Plants
 
SolarOnCare, H-Energy's solar power generation asset management platform, is joining forces with three universities—POSTECH, KAIST, and UNIST—to host a competition for university and graduate students nationwide. The practical competition utilizes actual solar power plant data to compete on methods for improving virtual power plant (VPP) operational efficiency, with the distinctive feature of consolidating and expanding the scale of two existing competitions.
 
SolarOnCare announced on the 22nd that it will jointly host the "Renewable Energy Generation Prediction Competition" with POSTECH, KAIST, and UNIST.

SolarOnCare has annually co-hosted a "Big Data Challenge" with POSTECH's Open Innovation Big Data Center (OIBC), and this year will integrate it with the "AI & Data Science Competition" of the three-university alliance.
 
Total prize money is 17 million won.

Awards will be given to a total of 17 teams: △Grand Prize 1 team (5 million won) △Excellence Award 2 teams (1 million won each for undergraduate and graduate students) △Merit Award 4 teams (1 million won each) △Encouragement Award 10 teams (400,000 won each). A separate undergraduate Excellence Award category was newly established with 1 million won in sponsorship from Omelet, a startup founded by KAIST Industrial Engineering faculty members.
 
Participants must, based on data from 253 actual solar power plants, optimally configure two aggregated resources with facility capacity of 20MW or higher and develop an algorithm to predict day-ahead generation.

The evaluation criterion is the settlement amount calculated by subtracting the generation prediction error penalty from power sales revenue, with teams that maximize this amount gaining an advantage.
 
A SolarOnCare official stated, "Not only the prediction accuracy of individual power plants, but also how resources with different generation patterns are configured affects the overall settlement results. We expect diverse aggregated resource configurations and predictive strategy algorithm ideas to be discovered."
 
Competition applications will be accepted from October 1 to 30 on the official competition website DataEn.

Participation is available as individuals or teams of up to 5 people, with an online briefing session held on October 12-13.

The main competition will run from November 2 to 13.
To request a correction, reply or follow-up report on this article, see how to file a request. Previously published statements are collected in corrections & replies.
명세환 기자
명세환 Reporter

Comments