Processing of forecast by ARIMA model on the weak electrical signals in Catharanthus roseus
Dong-Sheng Li · 2007
By taking an electrical signal in Catharanthus roseus as the time series and by using the model analysis of the autoregressive integrated moving average(ARIMA),an intelligent ARIMA forecasting system was constructed to forecast plant characters of signals de-noised by the wavelet soft-threshold backward.The weak electrical signal of C.roseus is Xt=2Xt-1+Xt-2+et-0.119 04et-1-0.361 83et-2-0.155 47et-3-0.363 66et-4.It is 15% at relative errors between the real and forecast values at backward 10 points of the electrical signal test in C.roseus,which shows it is feasible to forecast the plant electrical signal for a short period by the analysis of ARIMA.The forecast data can be used as important preferences for the intelligent automatic control system based on the adaptive characters of plants to achieve energy saving on agricultural production in the greenhouse and /or the plastic lookum.