Reliability of electrical machines and drives, part 2
Abstract
Stochastic models for forecasting daily and monthly peak demand in the urban electric utility system are investigated. An autoregressive integrated moving average model (ARIMA) was developed. Its efficiency was evaluated during one year. Results show a daily forecasting error of 2.5 % and a weekly forecasting error of 4.1 %. The average forecasting error during one year amounts to 3.7 %. It is concluded that ARIMA models are well suited for load forecasting.
- Publication:
-
NASA STI/Recon Technical Report N
- Pub Date:
- August 1983
- Bibcode:
- 1983STIN...8416481H
- Keywords:
-
- Autoregressive Processes;
- Electric Power Plants;
- Forecasting;
- Time Series Analysis;
- Cities;
- Reliability;
- Stochastic Processes;
- Utilities;
- Electronics and Electrical Engineering