Partial control of complex chemical plants.
Item
-
Title
-
Partial control of complex chemical plants.
-
Identifier
-
AAI9707065
-
identifier
-
9707065
-
Creator
-
Arbel, Arnon.
-
Contributor
-
Advisers: Reuel Shinnar | Irven H. Rinard
-
Date
-
1996
-
Language
-
English
-
Publisher
-
City University of New York.
-
Subject
-
Engineering, Chemical | Operations Research
-
Abstract
-
Two of the control problems common to many chemical processes are that they are nonlinear and, that there are more outputs to be controlled than manipulated inputs to control them with. In this latter situation it is desirable to choose a control structure that, while controlling only few variables at their set points, keeps the whole vector of specifications and constraints within prescribed limits. This is called Partial Control.;This research deals with of how to choose the basic control structure so that the two above objectives are met. First, the availability of manipulated variables has to be considered. In most cases this is a function of the design and is given for operation processes. Second, the outputs to be controlled with the available inputs need to be chosen. This work outlines a systematic approach for choosing the best control structure for a process. A set of criteria is suggested by which to compare the different possible structures.;The approach of partial control is demonstrated on a Fluidized Bed Catalytic Cracker (FCC). The FCC is one of the most important units in the refinery train. It converts heavy gas oil to lighter gasoline and gas products. It is a highly nonlinear process with a large dimension vector of specifications and constraints but very few manipulated inputs, hence a good candidate for partial control.;First, a new model of a generic FCC was developed. It was compared to other available models and commercial data. It was found to give good predictions of FCC operating conditions. Next, the nonlinearities of the process were studied. It was found that an FCC has, for a wide range of input conditions, both multiple steady states and input multiplicities. Several control structures that are in industrial use or discussed in the literature were investigated and evaluated based on the criteria suggested. This was done for both partial and complete CO combustion. It is shown that while for control at a given steady state linear methods are adequate, understanding the nonlinearities of the process is essential for choosing the basic control structure that will ensure stability and partial control.
-
Type
-
dissertation
-
Source
-
PQT Legacy CUNY.xlsx
-
degree
-
Ph.D.
