I have a dataset of 100 households' electricity consumption over 8760 hours. I have managed to plot the load over the year quite easily. However, I'm struggling to plot a graph showing changes over the 24 hours in a day.
householdData = pd.read_excel('/content/drive/MyDrive/Diss/cleanedInputData/100households.xls')
householdData
print(householdData)
Time Affluent Adversity Affluent Comfortable \
0 2019-01-01 01:00:00 0.254 0.244 0.155 0.215
1 2019-01-01 02:00:00 0.346 0.154 0.083 0.246
2 2019-01-01 03:00:00 0.309 0.116 0.085 0.220
3 2019-01-01 04:00:00 0.302 0.158 0.083 0.226
4 2019-01-01 05:00:00 0.181 0.171 0.096 0.246
... ... ... ... ... ...
8755 2019-12-31 20:00:00 1.102 0.263 2.157 0.209
8756 2019-12-31 21:00:00 0.712 0.269 1.409 0.212
8757 2019-12-31 22:00:00 0.398 0.274 0.073 0.277
8758 2019-12-31 23:00:00 0.449 0.452 0.072 0.252
8759 2020-01-01 00:00:00 0.466 0.291 0.110 0.203
I then set time as index by:
Energy= householdData.set_index('Time')
from this im able to resample to days like:
Daily_kWh= Energy.resample("D").sum()
Daily_kWh
I tried plotting the graph using
#load v time of year
fig, ax = plt.subplots(figsize = (16,5))
ax.plot(Energy, label='Hourly consumption',color = 'blue')
ax.set_ylabel('Hourly consumption [kWh]',size=15, color='green')
ax.set_xlabel('Time',size=15)
however this gives me a graph over the year. If i was to try find a graph over 24 hours for the households, how would i please do that. I've also tried aggregating all the loads into one hourly load but just can't seem to plot an hour graph.
Many thanks
Screenshots of codes.