Title :
A hybrid forecasting approach for HRG parameters based on output time series
Author :
Qi Ziyang ; Li Qinghua ; Yi Guoxing ; Fang Haibin
Author_Institution :
Space Control & Inertial Technol. Res. Center, Harbin Inst. of Technol., Harbin, China
fDate :
May 31 2014-June 2 2014
Abstract :
A hybrid forecasting model is proposed to predict the parameters of hemispherical resonator gyro (HRG) output time series. The hybrid forecasting methodology consists of an ARIMA model and a SVM model for the linear and nonlinear components, respectively. In the simulation, the proposed model is more accurate than single model in term of MAE and RMSE. Afterwards, the failure judgments are discussed, concluding the reason why performance deterioration leads to failure.
Keywords :
atmospheric techniques; autoregressive moving average processes; geophysics computing; support vector machines; time series; ARIMA model; HRG output time series; HRG parameters; MAE; RMSE; SVM model; failure judgments; hemispherical resonator gyro; hybrid forecasting methodology; hybrid forecasting model; nonlinear components; performance deterioration; Decision support systems; World Wide Web; ARIMA; HRG lifetime test; Hybrid forecast; Parameter extrapolation; SVM;
Conference_Titel :
Control and Decision Conference (2014 CCDC), The 26th Chinese
Conference_Location :
Changsha
Print_ISBN :
978-1-4799-3707-3
DOI :
10.1109/CCDC.2014.6852613
