MIMO Frequency Sampling Filters for Modelling of MIMO System Applications

Abstract

Abstract. In the modelling of a system based on approach, data acquisition is the first procedure that must be data acquisition process from data. This may lead to unacceptable computational time during the identification process. Raw data may also suffer severe noise disturbance high frequency region. In addition, bias estimatio considers direct identification from a closed problem, in this paper a introduced. Multi-input obtain only relevant model of the analyzed data. By performing this procedure, compressed, cleaned and unbiased data are produced. The efficacy of the MIMO frequency sampling filters was demonstrated by process data and steam generator plant data. raw data was successfully compressed and identification purposes with less computational time used to develop a model of the system, analysis, and also for developing a new Keywords: Data compression parametric model; system identification. IntroductionSystem identification is based on study and analysis of input and output data collected from a system. modelling of a system based on a system identification approach, data acquisition is the first procedure that must be carried out.acquisition process from a real system typically yields large amounts of This may lead to unacceptable computational time during the identification process. Raw data may also suffer severe noise disturbance, especially in high frequency region. In addition, bias estimation will occur if one only considers direct identification from a closed-loop system. To overcome in this paper a multivariable frequency sampling filter approach is input-multi-output (MIMO) raw data are analyzed in order to and meaningful parameters that describe the empirical model of the analyzed data. By performing this procedure, compressed, cleaned and unbiased data are produced. The efficacy of the MIMO frequency sampling demonstrated by compressing two sets of data: pH neutralization process data and steam generator plant data. The results show that the amount of successfully compressed and that the output was ready for identification purposes with less computational time, i.e. they could be further used to develop a model of the system, to conduct time and frequency response developing a new control system design.compression; frequency sampling filters; multivariable process system identification.System identification is based on study and analysis of input and output data system. To perform this procedure, a data acquisition process is unavoidable. Data acquisition from a real system typically yields large amounts time data. This may lead to unacceptable computational time during the identification process. Also, the raw data may contain complex system an extended version of the paper published by the authors in Int. Conf. Modelling ofElectronic Engineering, Universiti Sains Malaysia, School of Electrical & Electronic Engineering, Universiti Sains Malaysia, system identification carried out. The real system typically yields large amounts of This may lead to unacceptable computational time during ...