渐消递推滤波辅助光谱分析最佳自适应算法

应用科学学报 ›› 1994, Vol. 12 ›› Issue (2): 150-158.

渐消递推滤波辅助光谱分析最佳自适应算法

李志良¹, 余般梅², 钱锋³, 石乐明⁴, 潘忠孝⁴, 聂圣哲¹

1. 湖南大学四达应用化学研究所;
2. 国防科学技术大学;
3. 湖南大学;
4. 中科院化学冶金研究所

收稿日期:1991-07-15 修回日期:1992-09-06 出版日期:1994-06-30 发布日期:1994-06-30
基金资助:
国家开放实验室与自然科学基金及四达应用化学研究所基金

FADING RECURSIVE FILTERING AIDED SPECTROSCOPY An Optimal Adaptive Algorithm

LI ZHILIANG¹, YU BANMEI², QIAN FENG³, SHI LEMING⁴, PAN ZHONGXIAO⁴, NIE SHENGZHE¹

1. Hunan University STAR Institute of Applied Chemistry;
2. Chemical Metallurgy Institute of Technology;
3. Hunan University;
4. Chemical Metallurgy Institute

Received:1991-07-15 Revised:1992-09-06 Online:1994-06-30 Published:1994-06-30

摘要/Abstract

摘要： 研究了Kalman渐消递推最佳滤波辅助光谱辨析。基于Kalman滤波达到最优时其新息序列互不相关的性质,提出了一种新的渐消滤波——最佳自适应算法。通过在线自适应地调整遗忘因子从而使滤波(器)在存在模型误差或受到外扰时仍保持收敛性和最优性。用于重叠峰的辨识,能取得更好效果。

关键词: 光谱辨析, 自适应算法, 化学计量学, 多元分析, 渐消递推滤波

Abstract: The fading Kalman recursive optimal filtering is studied for spectroscopic resolution,Based on the nature of Kalman filter that the residual sequenees are not correlated when the optimal gain is obtained, a new fading filtering-optimal adaptive algorithm is proposed and utilized. Through the on-line and adaptive adjustment of the fading or forgetting factor,the convergency and optimality of Kalman filtering are improved using measured outputs or estimated results, even when there exist model errors and/or when the system is affected by unmeasurable external disturbances.The algorithm developed is applied to overlapped peak resolution with good results.

Key words: fading recursive filtering, Kalman filter, chemometrics, spectroscopic resolution, optimal adaptive algorithm, multicompo-nent analysis

李志良, 余般梅, 钱锋, 石乐明, 潘忠孝, 聂圣哲. 渐消递推滤波辅助光谱分析最佳自适应算法[J]. 应用科学学报, 1994, 12(2): 150-158.

LI ZHILIANG, YU BANMEI, QIAN FENG, SHI LEMING, PAN ZHONGXIAO, NIE SHENGZHE. FADING RECURSIVE FILTERING AIDED SPECTROSCOPY An Optimal Adaptive Algorithm[J]. Journal of Applied Sciences, 1994, 12(2): 150-158.