ABSTRACT
This chapter focuses on experimental testing and its application to the development of approximate process models. It leads to linear process models, although one of the methods presented can be used in nonlinear identification as well. The notion of linearity imposes certain restrictions on the testing procedures. The chapter moves on to step testing, a simple method for finding approximate process models. This will be followed by pulse testing which leads to transfer functions and Bode plots. The chapter discusses impulse response modeling and learns how impulse and step response models of multivariable systems can be developed. It takes up time series analysis and Pseudo-random binary sequence testing for single-input, single-output and multivariable systems. The chapter discusses of a random search procedure for process identification. If the mass and energy balances describing a polymerization system can be written with some degree of accuracy, the resulting set of differential equations is known as a state-space model of the system.
