DocumentCode :
3573951
Title :
Active data based Gaussian process models for nonlinear spatiotemporal systems
Author :
Pei Sun ; Lei Xie ; Junghui Chen
Author_Institution :
Nat. Key Lab. of Ind. Control Technol., Zhejiang Univ., Hangzhou, China
fYear :
2014
Firstpage :
6061
Lastpage :
6066
Abstract :
A new data-driven system identification method, called KL-GP, is proposed for spatiotemporal system. It combines Karhunen-Loève (KL) decomposition and Gaussian process (GP) models. As the nonlinear spatial-temporal spatiotemporal system has strong spatiotemporal characteristics, KL decomposition with good characteristics is employed for time/space separation and dimension reduction. Then the spatiotemporal output is expanded onto a low-dimensional KL space with temporal coefficients. GP models are employed to build up the temporal relation using these coefficients. In addition, a healthy spatial-temporal model that has accuracy predictions is always unknown in practice. GP provides an estimate of the variance of its predicted output. Using this characteristic, active data in the spatiotemporal system region can be found out for the model improvement. This enables the spatiotemporal system model to be updated without high computational demand. Simulation results of spatiotemporal system are presented to demonstrate the effectiveness of this KLGP modeling method.
Keywords :
Gaussian processes; identification; nonlinear systems; GP model; KL decomposition; KL-GP; Karhunen-Loève decomposition; active data based Gaussian process models; data-driven system identification method; dimension reduction; low-dimensional KL space; nonlinear spatiotemporal systems; spatiotemporal characteristics; spatiotemporal output; temporal coefficients; temporal relation; time/space separation; Computational modeling; Data models; Gaussian processes; Predictive models; Spatiotemporal phenomena; Training; Training data; Gaussian process model; Karhunen-Loève decomposition; Modeling; Spatial temporal system;
fLanguage :
English
Publisher :
ieee
Conference_Titel :
Intelligent Control and Automation (WCICA), 2014 11th World Congress on
Type :
conf
DOI :
10.1109/WCICA.2014.7053758
Filename :
7053758
Link To Document :
بازگشت