Gdz Elektrik Datathon 2023 - Time Series

📌 In this section, we will perform time series analysis by participating in the Gdz Elektrik Datathon 2023 competition.

Important Note: This notebook is a summary of the work done by many different people. I would like to thank everyone who has contributed to the work done and I hope it will be useful to everyone competing and good ideas will come out.

---

Business Problem

📌 It is expected that an "energy distribution" estimation approach will be developed for grid centers under the responsibility of Gdz Elektrik with the presented data set of the participants.

Dataset Story

Files:

• med.csv - Major Event Day dataset

• train.csv - Train dataset

• sample_submission.csv - Sample submission file

Descriptions:

• Major Event Day (MED.CSV): The day when the power outage time during the day exceeds the acceptable limit.Therefore, if there is an interruption, we expect the deviation of the estimated energy that day to be high. Because it could not be distributed due to the interruption.

• Dissipated Energy: Watt: Unit of active (useful) power. MWh = MegaWatt hour

• 85% of the distributed energy is for İzmir and 15% for Manisa.