Optimal measurement locations for parameter estimation of non linear distributed parameter systems

Brazilian Journal of Chemical Engineering

J. E. Alaña¹ 

[1] ,Universidad Del Zulia Facultad de Ingeniería Escuela de Ingeniería QuímicaManchester UK

关键词: Sensor placement;    Distributed parameter system;    Parameter estimation;    Sensitivity coefficients;   

DOI  :  10.1590/S0104-66322010000400015

来源: SciELO

【 摘 要 】

A sensor placement approach for the purpose of accurately estimating unknown parameters of a distributed parameter system is discussed. The idea is to convert the sensor location problem to a classical experimental design. The technique consists of analysing the extrema values of the sensitivity coefficients derived from the system and their corresponding spatial positions. This information is used to formulate an efficient computational optimum experiment design on discrete domains. The scheme studied is verified by a numerical example regarding the chemical reaction in a tubular reactor for two possible scenarios; stable and unstable operation conditions. The resulting approach is easy to implement and good estimates for the parameters of the system are obtained. This study shows that the measurement location plays an essential role in the parameter estimation procedure.

【 授权许可】

CC BY   
 All the contents of this journal, except where otherwise noted, is licensed under a Creative Commons Attribution License