一种新型GM(1,1)-AR预测模型在陀螺漂移中的应用

Application of a New GM(1,1)-AR Model in Gyroscope Drifting Tendency Prediction

为了对陀螺漂移趋势进行更有效的预测,提出一种基于小波分析的新型GM(1,1)-AR时间序列预测算法。该算法通过运用小波分解算法将陀螺漂移数据中的趋势项和随机项进行分离,然后分别运用GM(1,1)模型和AR时间序列预测模型对趋势项和随机项进行预测,最后用小波重构算法得出最终的预测值。给出了一种算法及具体步骤,最后用某型导弹陀螺漂移数据进行仿真实验,以检验这种算法的有效性和可行性,结果表明这种预测算法应用于陀螺漂移趋势预测是可行的。

In order to predict the gyroscope drifting tendency more efficiently, a new GM （ 1,1 ）-AR time series prediction algorithm based on the wavelet analysis is proposed. The algorithm separates the tendency part and random part of the gyroscope drifting data by using the wavelet decomposing algorithm. Then the tendency part is predicted by the GM （1,1） model, and the random part is predicted by the AR time series forecasting model. The final prediction result is the superimposition of the respective prediction by using the wavelet reconstruction algorithm. The algorithm is presented in detail herein. In order to verify the validity and feasibility of the algorithm, simulations were made by using the gyroscope drifting data of a certain type of missile. The results show that the proposed method in feasible.