Time Decoupled Prediction Models for Wind Speed and Solar Forecast

  • Unique Paper ID: 151371
  • Volume: 7
  • Issue: 12
  • PageNo: 546-551
  • Abstract:
  • With the increased necessity to integrate renewable energy into the existing power system network, methods aimed at optimizing the gap between demand and supply become important. This has drawn the interest of utilities towards developing and using state of the art forecasting techniques to predict wind speeds and solar irradiance over a wide range of temporal and spatial horizons due to their high dependence on local meteorological conditions, so as to more accurately forecast and deal with the variable power output from these plants. An efficient model that can predict wind speed and solar irradiance has become paramount. The paper has used a time series decoupling strategy to enhance accuracy of prediction. Instead of treating the data as a single time series, it is split into multiple time series in which each time series carries the historical data at a particular time of the day being considered. This has been applied to both solar irradiance and wind speed. By exploring conventional statistical models like Auto Regressive Moving Average (ARMA) and Auto Regressive Integrated Moving Average (ARIMA) with the use of time series decoupling methodology, we were able to perform a comparative analysis to assess the impact of the strategy as a singular variable that affects outcomes.

Cite This Article

  • ISSN: 2349-6002
  • Volume: 7
  • Issue: 12
  • PageNo: 546-551

Time Decoupled Prediction Models for Wind Speed and Solar Forecast

Related Articles