y*********m
发帖数: 33 | 1 大家给些思路吧，如果能帮我回答，我将感激不尽，真的急啊。谢谢了
A data file contains time series energy consumption data for a home over a
one month duration sampled at 1 second granularity
Format - Each line contains a tuple , like
below:
1312182000,1322
1312182001,1322
1312182002,1328
...
The timezone is PDT and the energy readings are in Watts.
The Data:
This is aggregate energy consumption data for one home that contains the
following main appliances,
* Central AC 1 - The most common repeating pulse. (At about 2.5 KW
amplitude and a width of about 10 minutes)
* Central AC 2 - Another repeating pulse but less frequent (At about 4
KW amplitude and > 30 minute width)
* Pool Pump - Runs for a duration of about 3 hours at 1.5 KW amplitude.
Starts at the same time everyday.
* Refrigerator - This is the smallest amplitude repeating pulse at < 200
W
If you are plotting the time series you should be able to spot all of the
above 4 quite easily.
The Challenge:
Process the above data to extract the energy consumption time series for
individual appliances listed above.
Provide a summary detailing,
* Techniques used (Brief note desribing why)
* Insights acquired while working on this data
* Interesting visualizations / observations
* Computational complexity of your technique(s)
* How would you solve this if you weren't provided the appliance
descriptions for the home? |
|
