BRAIN Hackathon: Energy Forecasting

BRAIN NTNU inviterer alle studenter til AI-Hackathon!

Arrangementet har vært.

“Energy is the key to a modern life. In industry, we try to predict how much we
produce. The objective is to establish a short-term (48h) forecasting system of the
wind energy production of a wind farm. There is a constant flow of incoming data
from wind turbines. Weather and sensor data are of key importance. The first
dataset will contain six months of historical weather information from the area of
windmills along with the turbine sensor data and past values of output power. To
prepare the dataset, you will need to treat the missing data, merge files or treat
misleading columns. The main feature of energy production forecasting system is
its flexibility. To simulate the real-life situation, we have added a gamification
challenge. The data will flow out in portions and you will start receiving additional
month of data within few hour intervals. As the new data portion arrives, you need
to prepare the new data, train and predict the turbine energy production for the
upcoming 48h window. Wind farm and electricity producers are required to
nominate upcoming production. Both underestimated and overestimated results
have an impact on the energy market. Therefore, the closest estimate of production
will be scored, and the scoreboard will tell if you are on the right track, so the next
portion of incoming data gets treated accordingly.”

Konkurransen starter fredag 6. mars og avsluttes på søndag 8. mars. Equinor har utarbeidet oppgaven, og vi kan friste med solide pengepotter til de 3 beste lagene! Førstepremien ligger på 15 000 NOK, andrepremien på 3 000 NOK og tredjeplass får 1000 NOK!

Tentativt program

16:15 – Velkomst, gjennomgang av oppgaven
17:00 – Konkurransestart
19:00 – Middag
23:59 – Dag 1 ferdig

08:00 – Oppstart dag 2
14:00 – Lunsj
19:00 – Middag
23:59 – Dag 2 ferdig

08:00 – Oppstart dag 3
11:00 – Oppgaven leveres inn, og lunsj
12:00 – Premieutdeling

Comments are closed.