Recursive Algorithm for Parameter Set Estimation

碩士 === 國立成功大學 === 化學工程研究所 === 82 === This thesis is concerned with the problem of set-membership identification. It refers to a class of recursive algorithms using ellipsoids to approximate the parameter set is formulated by a bounded-error...

Full description

Bibliographic Details
Main Authors: Feng-Yi Lin, 林鳳儀
Other Authors: Professor Shih-Y. P.;Professor Hwang-Chi
Format: Others
Language:zh-TW
Published: 1994
Online Access:http://ndltd.ncl.edu.tw/handle/24902252439522341206
id ndltd-TW-082NCKU0063034
record_format oai_dc
spelling ndltd-TW-082NCKU00630342015-10-13T15:33:32Z http://ndltd.ncl.edu.tw/handle/24902252439522341206 Recursive Algorithm for Parameter Set Estimation 參數集合估算之遞迴演算法 Feng-Yi Lin 林鳳儀 碩士 國立成功大學 化學工程研究所 82 This thesis is concerned with the problem of set-membership identification. It refers to a class of recursive algorithms using ellipsoids to approximate the parameter set is formulated by a bounded-error constraint.Three ellipsoid algorithms, including the optimal bounding ellipsoid (OBE),the optimal volume ellipsoid (OVE), and the fuzzy ellipsoidal set (FES), are rederived. It is shown that the ellipsoid algorithm with parallel cuts (EPC) and the OVE algorithm are the same. Due to the fact that the algorithms of EPC and OVE are mathematical equivalence, it can be concluded that the OVE and OBE algorithms are mathematically equivalent. Since the FES algorithm is developed very recently, it deserves detailed studies. The main characteristics underlying the FES is further explored. The above-mentioned ellipsoid algorithms are compared and interpreted geometrically. Moreover, a unified formulation for these algorithms is developed. Professor Shih-Y. P.;Professor Hwang-Chi 石延平;黃奇 1994 學位論文 ; thesis 122 zh-TW
collection NDLTD
language zh-TW
format Others
sources NDLTD
description 碩士 === 國立成功大學 === 化學工程研究所 === 82 === This thesis is concerned with the problem of set-membership identification. It refers to a class of recursive algorithms using ellipsoids to approximate the parameter set is formulated by a bounded-error constraint.Three ellipsoid algorithms, including the optimal bounding ellipsoid (OBE),the optimal volume ellipsoid (OVE), and the fuzzy ellipsoidal set (FES), are rederived. It is shown that the ellipsoid algorithm with parallel cuts (EPC) and the OVE algorithm are the same. Due to the fact that the algorithms of EPC and OVE are mathematical equivalence, it can be concluded that the OVE and OBE algorithms are mathematically equivalent. Since the FES algorithm is developed very recently, it deserves detailed studies. The main characteristics underlying the FES is further explored. The above-mentioned ellipsoid algorithms are compared and interpreted geometrically. Moreover, a unified formulation for these algorithms is developed.
author2 Professor Shih-Y. P.;Professor Hwang-Chi
author_facet Professor Shih-Y. P.;Professor Hwang-Chi
Feng-Yi Lin
林鳳儀
author Feng-Yi Lin
林鳳儀
spellingShingle Feng-Yi Lin
林鳳儀
Recursive Algorithm for Parameter Set Estimation
author_sort Feng-Yi Lin
title Recursive Algorithm for Parameter Set Estimation
title_short Recursive Algorithm for Parameter Set Estimation
title_full Recursive Algorithm for Parameter Set Estimation
title_fullStr Recursive Algorithm for Parameter Set Estimation
title_full_unstemmed Recursive Algorithm for Parameter Set Estimation
title_sort recursive algorithm for parameter set estimation
publishDate 1994
url http://ndltd.ncl.edu.tw/handle/24902252439522341206
work_keys_str_mv AT fengyilin recursivealgorithmforparametersetestimation
AT línfèngyí recursivealgorithmforparametersetestimation
AT fengyilin cānshùjíhégūsuànzhīdìhuíyǎnsuànfǎ
AT línfèngyí cānshùjíhégūsuànzhīdìhuíyǎnsuànfǎ
_version_ 1717766849732542464