基于小波支撑矢量机的陀螺仪漂移预测

Gyroscopic drift forecasting based on wavelet support vector machine

提出了基于小波支撑矢量机的导弹惯性器件故障预报方法。将子波核函数应用于支撑矢量机,得到小波支撑矢量机(WSVM)方法,用于某陀螺仪漂移时间序列预测中,预测精度优于基于传统核函数的支撑矢量机。通过仿真实验证明了算法的有效性和可行性。

In this paper, a new method based on wavelet support vector machine（WSVM） is put forward for fault prediction of missile inertial device. The wavelet support machine is obtained with wavelet kernel function, which is then used in time series forecasting for a certain kind of gyro. The accuracy of the prediction resuit is better than that of the Supporter Vector Machine （SVM） based on traditional kernel function. The simulated experiment proves that the algorithm is valid and feasible.