Regularized Least Squares Recursive Algorithm with Forgetting Factor for Identifying Parameters in the Grinding Process

This paper investigates a parameter identification problem in the grinding process. Due to the data saturated phenomenon and the ill-posed of parameter identification inverse problem, this paper presents a regularized least squares recursive algorithm with a forgetting factor (RLSRAFF), the basic id...

Celý popis

Uloženo v:
Podrobná bibliografie
Vydáno v:Journal of Mathematics Ročník 2022; číslo 1
Hlavní autoři: Yu, Yang, Deng, Rui, Yu, Gang, Wang, Yu, Yang, Guodong, Zhao, DaYong
Médium: Journal Article
Jazyk:angličtina
Vydáno: Cairo Hindawi 27.09.2022
John Wiley & Sons, Inc
Wiley
Témata:
Algorithms
Classification
Comparative analysis
Convergence
Error analysis
Error reduction
Grinding
Inverse problems
Least squares
Mathematical analysis
Mathematical models
Mathematics
Neural networks
Parameter identification
Process parameters
Random variables
Recursive functions
Regularization
Regularization methods
China
ISSN:2314-4629, 2314-4785
On-line přístup:Získat plný text
Tagy: Přidat tag
Žádné tagy, Buďte první, kdo vytvoří štítek k tomuto záznamu!
Popis
Shrnutí:This paper investigates a parameter identification problem in the grinding process. Due to the data saturated phenomenon and the ill-posed of parameter identification inverse problem, this paper presents a regularized least squares recursive algorithm with a forgetting factor (RLSRAFF), the basic idea of which is to combine the forgetting factor with regularization parameters. Moreover, based on RLSRAFF, this paper verifies the recursive calculation of criterion function, analyzes the effect of calculation error from the gain matrix and proves the convergence of the proposed algorithm. Finally, effectiveness of RLSRAFF is verified by simulation experiments and grinding data. Compared with other algorithms, RLSRAFF can give a more convergence rate to the real data and reduce the error from the true value.
Bibliografie:ObjectType-Article-1
SourceType-Scholarly Journals-1
ObjectType-Feature-2
content type line 14
ISSN:2314-4629
2314-4785
DOI:10.1155/2022/5188389