dineshresearch / ElectricVehicleDetection

From smart meter data (ie 30 min interval load) of 100+ houses, train classifier to detect if a specific house has an EV and when it is likely to be charging

Repository from Github https://github.comdineshresearch/ElectricVehicleDetectionRepository from Github https://github.comdineshresearch/ElectricVehicleDetection

ElectricVehicleDetection

Author: Robin Schucker date: 05/31/16

From smart meter data (30 min interval load over 2 month) of 100+ homes, train a classifier that detects if a house owns an EV and if yes when that EV is likeley to be charging.

Input:

EV_train.csv:

House_ID, Interval_1,Interval_2,Interval_3,Interval_4,Interval_5,...
1234567,1.013,0.215,0.217,0.217,0.217,...
1234568,0.22,0.22,0.215,0.215,3.612,....

EV_train_labels.csv:

House_ID, Interval_1,Interval_2,Interval_3,Interval_4,Interval_5,...
1234567,0,0,0,0,0,...
1234568,0,0,0,0,1,...

In label: 1 = EV is being charged, 0 = EV is not being charged

About

From smart meter data (ie 30 min interval load) of 100+ houses, train classifier to detect if a specific house has an EV and when it is likely to be charging


Languages

Language:Jupyter Notebook 100.0%