 |
International Society of Science and Applied Technologies |
| Data Sampling Intervals in Energy Research: An Exploration | ||||
| Author | Valentina Kuskova
|
|||
| Co-Author(s) | Dmitry Zaytsev
|
|||
| Abstract | The purpose of this paper is to evaluate various sampling intervals for obtaining real-life energy usage data. Current research does not explicitly consider sampling intervals, meaning the data are collected based on sensor settings, and are often stored in large files, increasing both the storage costs and the computational complexity of the models generated from these data. We demonstrate a statistical approach to sampling interval evaluation that can be implemented on a small sample of data, prior to sensor setup, which allows to select the most efficient sampling interval to be programmed into the sensor. If implemented, this approach will produce the sampling frequency that is fit for the needs of the data collection, reducing inefficiencies in data collection and storage while ensuring computational sufficiency for later model building. Our approach also demonstrates that random reductions in sampling frequencies can introduce statistical biases to the data, as some randomly created sampling intervals produced data that were statistically significant from the original.
|
|||
| Keywords | energy data, sampling frequency, statistical comparison | |||
| Article #: RQD2024-37 | ||||
Proceedings of 29th ISSAT International Conference on Reliability & Quality in Design |