Title :
Hessian Matrix Estimation in Hybrid Systems Based on an Embedded FFNN
Author :
Baek, Seung-Mook ; Park, Jung-Wook
Author_Institution :
Sch. of Electr. & Electron. Eng., Yonsei Univ., Seoul, South Korea
Abstract :
This paper describes the Hessian matrix estimation of nonsmooth nonlinear parameters by the identifier based on a feedforward neural network (FFNN) embedded in a hybrid system, which is modeled by the differential-algebraic-impulsive-switched (DAIS) structure. After identifying full dynamics of the hybrid system, the FFNN is used to estimate second-order derivatives of an objective function J with respect to the nonlinear parameters from the gradient information, which are trajectory sensitivities. Then, the estimated Hessian matrix is applied to the optimal tuning of a saturation limiter used in a practical engineering system.
Keywords :
Hessian matrices; differential algebraic equations; feedforward neural nets; gradient methods; nonlinear estimation; parameter estimation; time-varying systems; Hessian matrix estimation; differential-algebraic-impulsive-switched structure; embedded FFNN; feedforward neural network; gradient information; hybrid system; nonsmooth nonlinear parameters; optimal tuning; saturation limiter; second-order derivative; trajectory sensitivity; Eigenvalues and eigenfunctions; Hybrid power systems; Neural networks; Nonlinear dynamical systems; Optimization methods; Power engineering and energy; Power system analysis computing; Power system dynamics; Power system modeling; Tuning; Feedforward neural network (FFNN); Hessian matrix estimation; hybrid system; nonlinear parameters; optimal tuning; power system stabilizer (PSS); saturation limiter; Algorithms; Neural Networks (Computer); Nonlinear Dynamics;
Journal_Title :
Neural Networks, IEEE Transactions on
DOI :
10.1109/TNN.2010.2042728
