Scribd
Upload
108 views

Simulation Introduction

This document discusses different approaches to simulation of chemical processes, including sequential modular, simultaneous modular, and equation-solving approaches. The sequential modular approach involves breaking the process into separate computational modules for each unit operation. The simultaneous modular approach uses both exact models and linear approximations for each unit. The equation-solving approach formulates the problem as a set of nonlinear equations that must be solved simultaneously. Precedence ordering is used to break the equations into smaller subsets that can be solved sequentially. An example illustrates generating an incidence matrix and directed graph to perform precedence ordering on a set of process equations.

Uploaded by

anand_upadhyay_9
Copyright
© © All Rights Reserved
Available Formats
Download as PDF, TXT or read online on Scribd
108 views

Simulation Introduction

This document discusses different approaches to simulation of chemical processes, including sequential modular, simultaneous modular, and equation-solving approaches. The sequential modular approach involves breaking the process into separate computational modules for each unit operation. The simultaneous modular approach uses both exact models and linear approximations for each unit. The equation-solving approach formulates the problem as a set of nonlinear equations that must be solved simultaneously. Precedence ordering is used to break the equations into smaller subsets that can be solved sequentially. An example illustrates generating an incidence matrix and directed graph to perform precedence ordering on a set of process equations.

Uploaded by

anand_upadhyay_9
Copyright
© © All Rights Reserved
Available Formats
Download as PDF, TXT or read online on Scribd
You are on page 1/ 19

SIMULATION

Sequential modular approach


• This approach involves collecting equations and constraints for each
process unit into a separate computational subroutine or module.
• This concept is known as ‘unit computation’.
• Thus, each module calculates values pertaining to the output streams
for the given input conditions and parameters for that process unit or
equipment, irrespective of the source of input information or the sink
of output information.
• Using this concept, it is possible to create a library of modules and
use them for a variety of flowsheets
Sequential modular approach
Sequential modular approach
• In order to make it totally sequential, it is
then necessary to identify recycle loops of
units in a given flowsheet and to ‘tear’
certain streams.
• Tearing of streams implies assumption of
some value for that stream.
• For example, in Fig. , tearing streams 2 and
7, it is possible to make the calculations
sequential in order of the modules (1), (4),
(3), (5), and (2).
Sequential modular approach
• structural components:
1. A store of physical and thermodynamic properties for pure chemical species and their
mixtures.
2. A system-working database derived from the store as well as the problem input.
3. Individual module database derived from the working database.
4. Modules or subroutines for individual process units, computational procedures
including input/output for each module, and internally iterated variables, if any.
5. Flowsheet topology to be used by the system executive to order computational
sequence. This is usually defined by a ‘process matrix’ which specifies
6. standard unit modules for use in simulation of a particular process along with
identification numbers, positive for input streams and negative for the output
streams.
7. Routines for the recycle calculations and convergence analysis. Other mathematical
service routines.
Simultaneous modular approach
• In this type of approach,
• the unit modules are similar to those in the sequential modular
approach.
• However, the main difference is that for each unit, an additional
module is written, which approximately relates each output value by
a linear combination of all input values.
• Thus for the given input xi1, xi2, ..., xin for module i and equipment
parameters ui; the exact model for module i could be written as

•
• 11.1
Simultaneous modular approach
• and the linear models for module i could be represented by the following two
connecting equations:

11.2
11.3
• In the above equations, y denotes output values and β coefficients of linear
equations.
• The unknowns of the linear model ( β i j k) are determined by the simultaneous
modular approach described below.
• The flow diagram for the simultaneous modular approach is given
Comparision
• On comparing these two approaches, the sequential modular
approach seems easier to write; but in the simultaneous modular
approach, there is no decomposition required.
• Also, the convenience of the two approaches depends upon the type
of the process and the nature of recycle streams.
The Equation-solving Approach
• An unconstrained material balance problem is one in which all the
parameters are assigned known values.
• These parameters constitute the set of natural simulation
specifications, namely, the species, or stream split fractions,
stoichiometry, and key reactant conversion for all the reactors and the
species flow rates in all the external process streams.
Precedence-ordering of Equation Sets
• In general, the steady-state simulation of a process requires solving a
set of nonlinear simultaneous equations expressed in the form
fi ( x , u ) = O , i € In (11.4)
• in which x is a vector of n unknown state variables and u a vector of r
known decision variables.
• In an iterative procedure to solve Eq. (1 1.4), one has to start with an
initial vector xo and generate successive approximation to it xl, x2, ...,
so that the solution is obtained within a prescribed accuracy.
• The major drawback in most iterative techniques is that the whole set
of n unknown variables are to be iterated upon, which requires an nth
order convergence matrix. When n becomes large with the growing
complexity of the process, simulation needs larger storage space and
excessive computer time.
• It is, therefore, desirable to break up a large system of equations into
smaller subsets, which only need to be solved simultaneously.
• A systematic strategy to identify such subsets and to arrange them in
a proper sequence is known as precedence ordering.
• The equations are arranged in a certain order by first arranging them into
diagrams. The first of them is the Boolean ‘incidence’ (or ‘occurrence’)
matrix, in which the rows correspond to system equations and the columns
correspond to a system variable.
• Then the system is represented by a ‘digraph’ or directed graph or its
associated ‘Boolean adjacency matrix’.
• The output set assignment is not unique. Nevertheless, same results for
irreducible sets and precedence ordering are obtained.
• In this adjacency matrix, the rows correspond to the vertices from which
the flow originates and the columns correspond to the vertices to which
the flow is directed.
Example 11.1
• Find (a) the incidence matrix, (b) the digraph, and (c) the associated
adjacency matrix for a process represented by a set of the following
equations in functional form: f1 (x1,x 2)= 0,f2(x 4) = 0,f3 (x3,x 6) = 0,f4
(x4, x5) = 0,f5 (x1,x 6) = 0, & f6(x2, x3, x5) = 0.
• Solution
• (a) Incidence matrix In the incidence matrix, the process functions
and their relationship with the process variables are represented as
rows and columns of the matrix, respectively, using the binary coding
(Lea, 1’s and 0’s).
(b) Digraph In a digraph, the flow of information (i.e., the sequence of
calculations) is given. From the equation f2 (x4) = 0, the value of x4 can
be calculated.
• Using this x4 and the equation f4 (x4, x5) = 0, the value of x5 can be
obtained.
• Using this x5, and with an assumed value of x3, the equation f6 (x2,
x4, x5) = 0 gives the value of x2.
• Using this value of x2 and the equation f1 (xl, x2) = 0, the value of x1
could be computed.
(c) Associated adjacency matrix In the above digraph, the equations
are numbered as the nodes.
• In the adjacency matrix, rows correspond to the vertices from which
the flow originates and columns correspond to the vertices to which
the flow is directed; it can be obtained directly from the incidence
matrix as well.
• Rules of Boolean multiplication and union apply to these matrices.
REFERENCE

1. Process Plant Simulation, B.V. Babu

You might also like