Abstract The correlation analysis (CRA) theory is an important tool in the system identification field. Among its utilities, the correlation analysis enables the mathematical modeling of a process, through the estimate of its impulse response. This model can be used directly, as in some control algorithms, or, at least, can allow the attainment of important preliminary dynamic information that is valuable in constructing a parametric model of the system. This paper presents the mathematical development of the multivariable correlation analysis and its application to simulated and experimental data. Results show that the present method is effective in identifying an industrial polymerization reactor and overperforms the one commonly used, that suggests the use of the monovariable correlation analysis even for multivariable systems. A numeric analysis of the problem is also done and some guidelines are proposed to overcome possible problems.
[1]
R. K. Wood,et al.
Terminal composition control of a binary distillation column
,
1973
.
[2]
Chris Chatfield.
The Analysis of Time Series: Theory and Practice
,
1975
.
[3]
P. Young,et al.
Time series analysis, forecasting and control
,
1972,
IEEE Transactions on Automatic Control.
[4]
Lennart Ljung,et al.
System Identification: Theory for the User
,
1987
.
[5]
G. Stewart.
Introduction to matrix computations
,
1973
.
[6]
Babatunde A. Ogunnaike,et al.
Multivariable controller design for linear systems having multiple time delays
,
1979
.