Title
New results on Hessian matrices and stabilization for stochastic T–S models via line integral
Abstract
This paper is concerned with the stability analysis and stabilization for Itô stochastic T–S models via the line integral approach. Unlike the deterministic case, stochastic stability analysis of this model needs to handle the Hessian matrix of the line integral function. Since the Hessian matrix is dependent on the partial derivatives of membership-based functions, representing and estimating it are complicated and challenging and no relevant result has been reported so far. The novel representation of the Hessian matrix is skillfully provided which includes a part of the partial derivatives of membership-based functions. By giving an assumption on the partial derivatives of normalized membership functions, together with the property of matrix with rank one, an upper bound of the symmetric sum of partial derivative dependent part of the Hessian matrix is achieved. Thus a more general result of stochastic stability for the underlying model can be established by the line integral function. In terms of the obtained condition and a decoupling technique, a convergent cone complementarity linearization-like algorithm is provided for the controller design. Numerical examples are given to illustrate the effectiveness of the proposed scheme.
Year
DOI
Venue
2022
10.1016/j.automatica.2022.110337
Automatica
Keywords
DocType
Volume
Hessian matrix,Line integral,Quadratic optimization,Stochastic stability,T–S model
Journal
142
ISSN
Citations 
PageRank 
0005-1098
0
0.34
References 
Authors
14
3
Name
Order
Citations
PageRank
Shuigeng Zhou12089207.00
Y Han200.68
Bo Zhang3419.80