渐消卡尔曼滤波器的最佳自适应算法及其应用

AN OPTIMAL ADAPTIVE ALGORITHM FOR FADING KALMAN FILTER AND ITS APPLICATION

本文依据卡尔曼滤波器在使用最佳增益时,其余差序列互不相关的性质,开发了一种新的渐消滤波算法。该算法根据对象输出,在线自适应地调整遗忘因子,从而使滤波器在对象模型存在误差或对象受到外扰时,仍收敛并保持最佳性。该算法已应用于造纸机控制,取得较好效果。

A new fading filtering algorithm is developed based on the property of Kalman filter that the sequence of residuals is uncorrelated when the optimal gain is used. The algorithm adjusts the fading factor adaptively using measured outputs so as to improve the optimality and convergency of the Kalman filter when there exist model errors and/or when the plant is affected by unmeasurable external disturbances. The proposed algorithm has been applied to a paper-making machine control system and behaves satisfactorily.