Abstract | ||
---|---|---|
This paper provides new techniques to predict electric loads using a multiple linear regression (MLR) model, which adopts a statistical approach that assumes that past load and weather data can provide information for forecasting the target load. However, there are some application problems when the observed data is insufficient or the reference load deviates from the training data set. To solve these problems, we introduce new methods such as approximately adaptive searching and compensation. The results of case study show whether our new methods work well with real data. |
Year | DOI | Venue |
---|---|---|
2018 | 10.1109/SmartGridComm.2018.8587489 | 2018 IEEE International Conference on Communications, Control, and Computing Technologies for Smart Grids (SmartGridComm) |
Keywords | Field | DocType |
statistical approach,weather data,target load,application problems,observed data,training data set,multiple linear regression method,adaptive searching,MLR model,short-term electric load prediction | Training set,Electrical load,Smart grid,Computer science,Load modeling,Real-time computing,Load forecasting,Weather data,Linear regression | Conference |
ISBN | Citations | PageRank |
978-1-5386-7955-5 | 1 | 0.43 |
References | Authors | |
0 | 4 |
Name | Order | Citations | PageRank |
---|---|---|---|
Juntae Kim | 1 | 9 | 8.72 |
Seokheon Cho | 2 | 2 | 0.78 |
Kabseok Ko | 3 | 1 | 0.43 |
R. R. Rao | 4 | 1724 | 238.27 |