The Improvement And Application Of Time Series Models

Time series models are mathematical models of reflecting system dynamic structure and laws. However, time series models are not very satisfactory. In order to improve accuracy, according to the characteristics of the observed data, two times series models can be improved.Within a wind energy system, the wind speed is one key parameter. This paper proposes a improved time series model for long-term wind speed forecasting based on the first definite season index method and ARMA model or GARCH model. The simulation results show that the developed method is simple and quite efficient for wind speed forecasting compared with ARIMA model.Half-hourly electricity price in power system are volatile. However, the fluctuation depends on many factors. Therefore, it is difficult to use only one model. This paper proposes the second improved time series model that detaches high volatility and daily seasonality for electricity price of New South Wales in Australia based on EMD, Seasonal Adjustment and ARIMA model. The results demonstrate that the proposed model can improve the prediction accuracy noticeably compared with Seasonal ARIMA model.