coDEN: LUTFD2/(TFRT-3176) I r-87 I Fe85)

A SIMI\ON TUTORIAL
o

Karl Johan Astrm

,,|r}urtment of Automatic Control

Lund Institute of Technology
July 1985

A SIMT{OT{ TIJTORIAL
Karl Johan Astrm

r
l'

Department of Automatic Control

Lund Ins+iute of Technology
t
lt

l1

't

Revised edition, July 1985

Documcni ram

Department of Automatic Control

Lund Institute of lechnology

REPOR,T
Dte ol csuc

July

P.O. Box 118

5-227 00 Lund Sweden

1985

Documctt Numbe

CoDEN: LUTFD2/(TFR-3176)/t-82/(1e85)
Supewsar

Author(s)

Karl Johan ,"ttS p o ns o

rng or;'enisa oa

Title and subtitle

A Simnon Thtorial

Abstract

The purpose of this report is to give a tutorial introduction to the simulation language SIMNON

\
t'

Key words

Clas-"ifcair syetem

aad/or inilex tcrms (i q)

,i
Srrpple.urenary

,,

bibliographicalnfomaton l t '

l.

ISSN and hey ttle

Languagc

English

ISBN
Numbe of pzgec

Recpen't n.oles

87

Seurty clssfrcton

The rcpoft m4y be odered from the Department oi automtc control o bottovcd through the Unventy Lbrary 2, Box 7070,
5-221 03 Lnnd, Sweder, Telex: 33216 tlubbis funil"

Contents

1. Introduction

2. T}re Conceptual Framework

3. Differential Equations 10
4. Difference Equations 22
5. Combination of Systems 29
6. Advanced Features 35
7. Implementation 42
8. References 45

A
Appendix B
Appendix C
Appendix D
Appendix E
Index 81
Appendix

- Syntax for Simnon Commands 47

- Syntax of System Descriptions 59
- Standard Systems 61
- Intrac and Simnon Commands 74
- Macros for Generating the Figures

79

l
, l'

,r*;

1. lntroduction

Simnon is a special programming language for simulating dynamical systems.

Systems may be described as ordinary differential equations, as difference
equation or as combinations of such equations. Models of this type are common
in mathematics, biology, economics and in many branches of engineering. Simnon
requires a computer with a graphics terminal. The results are displayed as

curves on the terminal. The language has an interactive implementation which

makes it easy for a user to work with the system. Simnon may be used in a very
simple way to find solutions to difference or differential equations. This requires
only six commands. There are 38 additional commands in the system. They also
allow optimization, introduction of experimental data and parameter fitting"
The purpose of this report is to provide an introduction to the simulation
language. The conceptual framework is first described in Chapter 2. Thre report
then proceeds by examples. Chapter 3 describes how to solve simple differential
equations. Solution of difference equations is described in Chapter 4. Chapter 5
describes simulation of more complicated systems which are obtained by
combining subsystems described by differential or difference equations. It is
useful to note that all of this can be accomplished by about a dozen commands.
Some advanced features are described in Chapter 6. A few remarks on the
implementation are given in Chapter 7. Eh chapter is provided with exercises.
Do not forget to experiment and test at a terminal as you prog."r, with the
l.*"ys' type HELP and that there is also a
reading. Remember that you
""r, t
manual which gives a detailed description of the language constructs. Also
remember that a good way to learn the language is to start by learning how to
master a few commands and expana
l1

f f
n

ie vocabulary gradually.
,

2. The Conceptual Framework

Simnon

is an interactive

language

for

simulating dynamical systems. The system

may be described by ordinary differential equations,

or difference equations. It

is

also possible to simulate systems which consist of interconnected subsystems.

This is useful in order to structure a large system. Simnon may also be used for
other purposes e.g. to graph functions to fit models to data etc.

The simulation language has facilities to edit system descriptions, to integrate

differential equations, to store and retrieve data, to show the solutions as graphs,
and to change parameters and initial conditions.

DESCRIPTIONS OF DYNAMICAL SYSTEMS

To use Simnon it is necessary to have a basic understanding of dynamical

systems. In particular to be familiar with the notions of input, output and state.
The generic form of a continuous time system is

$f = r(*, .',)
Y = g(x, u)

(2.11

where u is a vector of inputs, y a vector of outputs and x is the state vector. A

system like (2.1) is specified as a CONTIITIUOUS SYSTEM in Simnon. The analogous
form for a discrete time system is I

*(.r*r) = r(x(t), u(t)),

v(tr) =s(x(tt) ,u(t))., k-1,2,..

(2.21

.t ,i

A system like (2.21is specified/ as a'ISCRETE SYSTEM. Simnon allows a system

or a subsystem to be described by either of the forms (2.1) or (2.2). lt is also

possible to have interconnected systems where each subsystem has the form (2.1)

or

(2.21. Connections are described as a CONNECTING SYSTEM.

INTERACTION PRINCIPLES

to the user via a graphical screen which can show

curves, text and numbers. It is also possible to get a hard copy of a picture and
Simnon gives information

to list system descriptions and data. Simnon receives information from the user
by commands from the keyboard. The commands have the forrn
CHND

argl

arg2

where CMND is the name of the command and argl, arg2, etc. are the arguments.
The name is a combination of up to eight characters. The arguments may be
identifiers or numbers. Spaces are used as separators. A command is terminated
by carriage return (CR).
Default values

It is

deslrable that commands are both short and flexible. One possibility to
achieve these conflicting goals is to allow variations of a command which are
selected by the arguments. A description of the simulation command illustrates
the idea.
The simulation command

The different forms of the command SIMU which executes a simulation of a

system are illustrated in the syntax diagram Fig. 1. The diagram implies that any
form of the command which is obtained by..traversing the graph in the directions
of the arrows is allowed. For example the command

SIHU O 1oO

,,

t
' l'
simulates a system from time 0 to time 100. If we want to repeat the simulation
second time with different parameters it suffices to write

SIHU

"4

The previous values of the afBrr6ets, i.e. O and 100, are then used. Most
commands are provided with default arguments. These arguments are then used
unless otherwise stated,
7

Time increment

Stop time

@
SIMU
T

CON

Filename

Store increment

Fiqure 1. Syntax diagram for the command SIMU.

It follows from Fig. 1 that start and stop times and the initial time increment may
be specified. It is also possible to mark curves by the optional argument MARK.
A simulation may also be continued by using the end conditions of a previous
simulation as initial values. This is done by the command option CONT. The
results of a simulation may also be stored in a file. The time spacing between the
stored values is specified by increment.
The syntax

for the simulation

command may also be described as follows

increment> ll
t <start time> <stop time> t ((increment>
ilename>
ll
t -{coNT l}IARK}l [/<f
t
This description is called the Backus-normal form (BNF), or the Bachus-Naur
form. (...> denotes an argument i.e. a number or an identifier. [al a2...1 denotes
optional arguments which may be omitted and {al1"21...} denotes that one of the
SIHU

alternative arguments must be chosen. An asterisk (*) after an argument denotes

ny command is obtained by typing the
that it may be repeated. ffre synta|,fo
'rl
command HELP followed by the name of the command. The syntax is then shown
in the Backus-Naur notation. The syntax of the commands is given in Appendix
A

4,

l', ".

EXERCISES

1. Learn how to log in and log out of your computer system and how to get
access to mass storage areas. Get help from your systems manager if
necessary. Document what you are doing so that you can repeat it on your
own.

2. On the Vax implementations you start Simnon simply by typing Simnon.

Simnon answers by the prompt >. You can exit Simnon by typing STOP.
Practice this.

3. Try to find out about Simnon by using the command

HELP.

4" Use the HELP command to find out how the command AXES works.
with the AXES command to find out how it works.

' It

tl

F;<periment

tt,

Itn

=-

3.

Differential Equations

Solution

of ordinary nonlinear differenar

is a simpre apprication of
simnon' In such apprications
simnon can be viewed simpty
as a carcurator for
differential equations' There
are two minor differences compared
to a calcurator.
Simnon can dispray curves.
rn a carcurator a function is
activated by pushing a
dedicated key' In simnon
functions are activated by typing
a command on the
keyboard' How sirnnon may
be used to generate sotutions
to an ordinary
differential equation is illustrated
by an example.
equations

THE PROBLEM

::i::,,tt

we would like to know the character

or the sorurion to the van der

.2

#.^2-b)T+y=o

(3.1)

for different initial conditions and

different values of the parameters
a and b.
van der pol equation is a model
for an elqctronic oseillator.
The path toward: s the solution
tolthe Poblem can be divided
steps.
, lt

i.
2.

The

in the following

Enter system descriptions

Simulate

nditions

il a satisfactory result
different steps will now be described
in some detail.
10

is

obtained. The

continuous system \|DPOL

"The van der Pol equation

state x

der dx dy
dY=x

dx=a*x* (b-y*y)-y
a:1
b:1
END

Listine

1. A Simnon

system for Equation (3.2)

ENTER THE SYSTEM DESCRIPTION

first rewritten in the standard state space form (2.1). Since

the equation (3.1) is of second order an extra variable is introduced. The
The equation (3"1) is

equation (3.1) can then be written

as

dv

=*
(3.21

The equations are now in standard state space form, which is the format
necessary to use Simnon" A file which describes the system should now be
prepared. This file which is labeled VDPOL is listed in Listing 1. The first line
indicates that it is a continuous time system with the name VDPOL. The state
variables x and y and their derivatives dx and dy are declared. The differential
equations are then defined. Notice the strong similarity with (3.2)" Finally values
separates between @g!g!
are assigned to the parameters a and b.
-Simnon
and varia. Parameters may be assigned values in a system'description using
the notation ':' for assignment. Parlmetqrrvalues may be reassigned interactively
using the command PAR. Variables e efined using the notatiol '='.

The file can be edited using any editor you are familiar with. Assuming the
standard editor you type EDIT VDPOL,.T if you are in VMS and $EDIT VDPOL.T if
you are in Simnon. In the Vf/VlfS
system all Simnon system files have the
t /
l1 r
extension'.T'. This also appliesTto Mdcro files (Chapter 6). There is also a simple
line oriented editor built into Simnon which can be used to enter the file. This

1L

editor is invoked by the command

EDIT <filename>

where the argument <filename> is an identifier i.e. a letter possibly followed by

letters or digits. The facilities in the editor are described in Appendix A. The
syntax of the system descriptions is given in Appendix B. The command
LIST <filename>
lists a system description on the terminal.

SIMULATION

To run a system

it must first

be activated. This is done by the command

SYST VDPOL

lf there are any errors during

error

is given and the

system enters the Simnon editor so that the eror may be corrected. If you want
to use another editor simply type LEAVE to exit Simnon's editor.
the compilation an

message

If we would like to see the solution curves as they are integrated we must first
draw axes on the screen. This is done with the command

AXESH020V-66
which means that the ranges of the horizontal (H) and vertical (V) axes are
chosen as (0,20) and (-6,6).
The command
PLOT

instructs the program to plot the vniables' x and y as functions of time.

' l'

'

To perform a simulation it is necessary to give appropriate initial conditions to

the state variables. The command

INIT x: 1
assigns the initial value 1

,l
to

tf

set to zero if no assignments are

simulation. The command
L2

x. lnitial values are automatically

made. We are now ready to perform a

th t t'atp variable

ot
!

c
c

H
!
c)

J
L

t
!

-2

a!

o
a,

.;C

-4

(,
+t
.l

+)

al,

-6
0

Timc

Figure

2.

t0

Simulation of the van der Pol equation

values x(0)=1 and Y(O)=g.

20

15

for a=l and b=l with

initial

SIHUO20-HARK
activates a simulation from time O to time 20, and the result shown

in Fig. 2 is

obtained. The argument -MARK implies that the curves are labeled with integers
L,2u.. in the order they appear in the plot command.

HOW TO INTERRUPT

SIMULATIOfr

' I'

Some times when you make a simulation you

will find that the results are wrong

at the beginning. It is then useful to'be able to break the simulation immediately.
There are facilities for doing this in Simnon. The details are implementation
dependent On the standard V""!y4"rns the simulation is interrupted by typing
ltt

CTRL-C.

13

CHANGING PARAMETERS

Suppose that

it is of interest to explore how the character of the solution

is

influenced by the parameters a and b. The command

PAR

b:2

assigns the value 2 to the parameter b. The command

SIHU

now generates a new simulation. Notice that it is not necessary to specify any
arguments in the command SIMU. The previously given values 0 and 20 are
automatically used as default values. The use of default values simplifies the user
interaction considerably.
continue to explore how the solution is influenced by the
parameters of the solution by a repeated use of the commands PAR and SIMU.

It is now easy to

Curent values of all states, derivatives, variables and

parameters may

be

displayed. The command

DISP

displays all

curent

data. Selective displays

the variable y is done by the command

of the parameter a, the

state

and

DISPaxy
The display may also be directed to the line printer.
how the command DISP works is to type

simple way

to find out

HELP DISP

The help function can be applied to all commands

I'

, l'
STORING AND EDITING RESULTS OF

SIMULATION

some results, to compare results from different

simulations and to 'plot different,'J-ta/e- variables in different diagrams. Suppose for
|*,
the results for the parameter sets a=1,
example that we would like to compare
"
b=l with a=!, b=2. Two data files are first generated. The command

tt may be useful to store

T4

PAR
PAR

a:1
b:1

sets the parameters. The command

STORE

tells that the state variables x and y should be stored. The command
srHU/B1

then performs a simulation and stores x and

is set to 2 by
PAR

y in a file

called 81. The value

of

b:2

and the command

srMU/82
simulates and stores the results in file 82. The command

SPLIT

splits the screen into two windows. The command

ASHOI{

x/82

plots the variable

x from file 82 in the first window using automatic

scaling. The

command
SHOI{

x/81

plots the variable x from file 81 in the same window. The commands
ASHoll y/82

SHorl vlBL
plots the variable y frorn files 82 and 81 the second window. To document the
results it is useful to generate a hardcopy of the curves obtined. The details
depend on the hardware. On a norrLal insNallation the command
t
, l'
HCOPY

sends a copy of the picture on thg screen to the plotter queue. The results
shown in Fig" 3 are then obtained. Th command

sPLIT 1

f,
,",
l'*

clears the screen and resets plotting to one window which covers the full screen.

t5

x
o

.;al

3!

+t

lt

{J

U'

-6.
0

Tine

10

15

20

10

15.

20

a,

It
.;c

o
o
n

+t

4J
U)

-3.
0

Timc t

Fisure

3.

Solution of the van der Pol equation

for b = 1 and b = 2.

PHASE PLANE PLOTS

For two-dimensional differential equations it is useful to visualize the results

as

phase plane plots. This may be done simply by

AXESH-44V-33
SHot{

y(x)/41

which generates Fig.

4.

+
' l'

GENERALITIES

The general way of using Simnon as a 'calculator for differential equations" will
now be given. The generic for?'{f drsystem description is given in Listing 2. Trre
system description starts with CONTINUOUS SYSTEM <Identifier>.

It is terminated

by a line which contains END. An identifier is a a sequence of letters and digits

16

-1.
'l
@

.o

.;o

-2.

+t

+t

al,

-4"

-2

Statc variable
Figure 4.

Phase plane plot of the van der Pol equation

Y(o) = o.

CONTINUOUS SYSTEH

for a = 1, b = 1, x(0) =

1,

<Identif ier>

differential

"GeneraI
equation
Comm en t
state <Identifier>. . .<Identifier>
der <Identifier>. . .<Identifier> Declarations
time <Identifier>
computation of auxiliary variables
computation of derivatives *
Bod
v

parameter assignment

initial value assigrunnt

END
, t, I
Listine

2. Generic form of system description for simulation of

differential

equations.

'4

,t

fi,,!'|
lt,i

where the first character mus tbea letter. Both upper and lower case letters
may be used although the compiler does not distinguish between them. The

T7

system description has two parts, a declaration and a body

There are three types of declarations. A time variable may be declared for
simulation of time varying differential equations. This is done by the keyword
TIME followed by an dentifier. The state variables and their derivatives are
declared by the keywords STATE and DER followed by a list of the state variables
and the derivatives. associated by their sequential order in the lists.
The body of the system description specifies the derivatives of the state variables
in terms of state variables and parameters. Auxiliary variables may also be used.
The body also contains assignment of parameters and initial values. The order of
the statements in the body is unimportant. A variable may only be defined once.
Expressions and operators
The expressions available in Simnon are similar to those in a procedural language
like Algol or Pascal. An expression may be a string, a numeric constant or a

variable. It may also be combinations of variables, operators and functions.

Conditional expressions of the form F...THEN...ELSE are also permitted. Simnon
has arithmetic, relational and logical operators. All variables are floating point
numbers. The numbers are written in the conventional way as 4, 1.1 or 6E7. The
result of boolean expressions is 1.0 if it is true and 0.0 if it is false.
Arithmetic operators

The arithmetic operators are addition, subtraction, multiplication, division

and

exponentiation. They are denoted as

+-*

respectively.
Relational

operators

+
, l'

There are three relational operators, namely equal, greater than, and smaller
than. They are denoted bY

=><
Logical operators

t,
| ^, ' ',

The logical operators are AND, OR and NOT.

18

Functions

The following functions are available:

abs

(x)

absolute value
x<0
x=0
x>0

sign(x)

{l

int

largest integer less than x

xmody
Iargest of x and y
smallest of x and y
sguare root of x, x)0
exponential function
natural logarithm of x
logarithm (base 10) of x
sine of x (x is in radians)
cosine of x (x is in radians)
tangent of x (x is in radians)

x)
xrY )
X,Y

mod
max

min(x,y
sqrt (x)
exp (x)

Itr ( x)
log (*)

sin (*)
(*)
tan (*)
arcs in (")
erccos (*)
cos

atan (x)
atan2 ( x,Y)

sinh(x )

cosh (x )
tanh ( x )

arcsine
arccos

arctangent
arctangent
hyperbolic
hyperbolic
hyperbolic

of x result in (-n12, "l 2l

of x/y result in (-n, r)
sine
cosine
tan

SUMMARY

It has been

demonstrated that Simnon may be used

to

generate solutions to
ordinary differential equations in a very simple way. To do this the differential
equations are first written as a system of first order equations like
dx
Ar

= f (x,t).

, l'

continuous time system is then generated in Simnon by declaring the state

variables and the derivatives. The fuhction f, which defines the right hand side of
the differential equation, is ,lthen introduced using ordinary mathematical
,,
expressions and assignments of pahmeters.
There is no vector notation so all
tl

equations must be written

I t.

in scalar notation. Any editor may be used. There

is

also a special editor incorporated into Simnon, which is invoked by the command

19

EDIT. The command LIST can be used to list files.

six basic commands: SYST, AXES, PLOT, IIJIT,

A simulation is executed using

P.A'R and SIMU.

In order to edit, manipulate and document results from several different

simulations it is, however, also useful to use six additional commands, namely
STORE, SHOW, DISP, SPLIT, AREA and HCOPY.

The HELP command is useful in order to see what the commands do.

EXERCISES:

Learn how to use an editor on your system. Practice by editing the system
in Listing 1.

2.

Learn how to enter and exit from Simnon. Use the command LIST to list the
system you have edited on the screen and on the line printer. (Note in the
Vax implementation you may invoke a program called PRG from Sir:nnon by
typing $PRG without leaving Simnon.)

Repeat the simulation described

Change parameters

in this chapter on your own.

with the command PAR and initial conditions with the

and
investigate how the solutions to the van der Pool
command INIT
equations change.

5.

Experiment with the commands AXES, SPLIT, SHOW, ASHOW and AREA.

6.

Use the HELP command to investigate the basic simulation commands AXES,
PLOT, INIT, PAR, SYST and SIMU.

7. Introduce a formal error in the program in Listing I by changing the

assignment of dx to dx = ar,x*(b-y*y-y. Try to simulate the incorrect
program. (When an error is detected the command EDIT is automatically

executed. You may corect the mistak using the line editor. The simulation
is then automatically continued when you exit the editor by iyping E" You can
also leave the editor by the cqnman{ LEAVE and use your favqurite editor

forthecorrection.)

8.

,It

Use the HELP command to investigate the auxiliary commands STORE, SHOW,
DISP, ASHOW, SPLIT and AREA.

Look at the commands LIST, LP eind HCOPY and learn how to get hardcopies
of listings and plots on yout,fsYftem.
r|
h,,!
/dejr
Pbl equation
10. Consider the following van

9.

t+ (v2-t) *Y=a
20

Investigate the limit cycles obtained

for

e = 0.05 and 0.993

( a(

1.

11. The following equations called the Volterra Lotka equations represent
simple model for the development of two competing species

dx

=(a_by)x

T=(cx-d)y
Make a Simnon program to study the equations. Start with the following
nominal values: a o 2.7,b = 0.7, c = 1 and d = 3.

72. A simple model for a satellite orbit is given by

d2"
dtz

la'12
ldtj
l-l=_-

& +-r
2

dt2

.2

dr d<p
dt dt =Q

where r is the radius from the center of the earth, g the azimut angle and k
a gravitational constant. Simulate the equations.

13. Learn how to interrupt a simulation on your installation.

L4. Simulate the differential equation
dx

="(v-*) ,
dy
y
dt = bx - - xz,

y(o) = -a

dz
T

zl9l=24

=-cz *xy,

x(o) =-a

where a = 10, b = 28 and c = 813, Look in particular at z as a function of x.

The equation, which is called the Lorenz equation, is an example of a
deterministic system, which has a very, irregular (chaotic) behavior.

\
, l'

.l

tf .
Itc l',

2t

4. Difference

Equations

There are also facilities for simulating difference equations

in simnon. This is
done analogously to simulation of differential equations.
An example is first given
and the general principles are then stated.

EXAMPLE

Consider the following difference equation

*k*1 = *k *. * xk * (l-xk) ,

k = 0,1,...

(4.1)

This is a simple model of a population dynamics. The variable

xu denotes the
number of individuals at time k in a normalized scale.
There is an equilibrium
value x=l at which the population remains constant. For
xno the population
increases with the factor 1+r in each generation.
For xu>l the population will
decrease' Assume that it is of interest to investigate
how the population changes
with time.

difference equation is characterized as a DISCRETE SYSTEM

description of a system which simulates the difference equation

Listing3.

in

Simnon. The
(4.1) is given in

The state variable x is declared in',rl"'

""1" way as for continuous time sysrem
using the srATE declaration. The declaration
NEW is used to declare the variable
nx, which denotes the new value of the,state variable. NEW
is thus the analog of
DER for continuous time systems. ,{ ,i

l',f'

when simulating difference equations it is necessary to provide

a mechanism for
making the state variables change at certain sanpling
times. For example the state
22

discrete system
"Simple model

state x

new nx
time k
tsamp ts
nx=x+rr x*

POPDYII

for population

dynamics

t-*)

15=t+1

x: 0.5

r:2

END

Listins

3. A Simnon

system for simulation of the difference

equation (4.1).

variable in equation (4.1) changes periodically at k = !,2, .... In the general case
there may, however, be irregular sampling. The mechanism used to describe this
in Simnon is to introduce a variable TSAMP which gives the next sampling time.

In Listing 3 the first assignment statement is simply a definition of the right hand
side of the difference equation (4.1). The second line: ts=k+l assigns the value
k+l to the TSAMP variable ts. This tells that the system will be activated next at
time k+1.

SIMULATION

Discrete time systems are

run in the same way as continuous time

systems. A

simulation is thus executed by the commands

syst
split
axes
plot
simu

popdyn

31
H080V01
x
0 80

X ,r
, l) t

This generates the uppermost curve in Fig. 5. The commands

axes

par r:2.7O

s imu
axes

par r:2.83

t f,,

t,,

simu

23

x
c

0.75

;a
f

CL

o-

0
0

20

40.

60

80

20

40

60

80

20

40

60

80

0.75

;
a

0.75

;
f

o-

0.

Gcneration

Fisure

5.

Simulation of population dynamics.

for r = 2.7O and r = 2.83, and plots the corresponding

curves shown in Fig. 5. Notice that the behaviour of the solutions change
repeats the simulation

drastically with moderate changes in the parameters"

GRAPHS OF

FUNCTIONS

.t

, l'
Simnon is conveniently used to obtain a graph of a function. Assume for example
that we want a graph of the polynomial

(*) = x(x+3) (x+2) (*-21(x-e)

.ti

This is accomplished by the follotr,f tyst"-.

24

150.

f (x)
100.

50.

-50

- 100

- 150
0

-4

2
x

Fieure

x(x+3)(x+21(x-21(x-3) generated by

6. Graph of the function f(*)

Simnon.

discrete system POL

time x
tsamp z

=(x+3)* (x+2)*x* (x-2)* (x-3)

2=*dX

dx:0.05
END

The commands
th

syst pol
store f
simu -3.6 3.6
ashow

, l'

t
I

will then generate Fig.

6.

"t

'f

Simnon is also convenientfy .e/tto generate level curves, field plots and
conformal maps. To illustrate such applications consider for example the problem

25

;t"jt"ot"t

the image of the lines arg

z

G(z) =

trr2 and arg z =

3rr.in

the conformar

", = "*=iy = "*("os y + i sin y).

This may be accomplished

by the s!stem

discrete system
time r

tsamp

El(pZ

fi=atfa*pi/ tg}
x=r*cos (fi )
y=r* s in (fi )

ReG=exp (x) * cos

(y)
ImG=1xp(x)*siniyj
s=r*dr
dr:0.01

pi: 3. 74t5926
alfa: g0
END

The commands

syst

EXpZ

axesh-11v0l.s
plot ImG(Rec)

srmu 0 3.14
par alpha: 13S
simu 0 4.44

then generate the graph shown

in Fig.

Z.

GENERALITIES

The generar form of a description

of a discrete'.rime system is given
in Listing 4.
It is analogous to the continuous
tim,
onlv
semantic
difference is thar
the sampting variabte TsAMp
must b:e :1t-
assjneC,Tlre
and that der is replaced
by new.
SUMMARY

Difference equa ions are

simutated in

rl, (r,n" way as differentiar equarions. The

SYSTEM dESCriPtiON
DISCRETE SYSTEM

26

iS

USEd iNStEAd

Of CONTINUOUS

SYSTEM.

1.5

Im0
L.25

0.75

0.5

o.25

-0.5

Fisure

7.

ReG

Graph of thermap of the rays arg

map G(z)

0.5

=e.

z = r.II

discrete system <identifier>

"general difference equation
state <identifier>...<identifier>
new <identifier>. . .<identifier>
time <identifier>
'
tsamp <identif ier>

computation of auxiliary variables

computation of new vafues of the states
update the variable Tgmpr
parameter assignmeni'

and arg

z = 3rl4 in the

comment

declarations

body

initial value assignment

END

Listins

4. Structure of Simnon /tF* for simulating difference

equations

ln
l't 'I

27

The simulation commands are the same as

for differential

commands AXES, PLOT, INIT, PAR, SYST and SIMU and

STORE, SHOW, DISP, SPLIT, AREA and HCOPY.

, lr

lr
|','"

28

equations, i.e. six basic

six auxiliary

commands

EXERCISES

1.

The difference equation

*r*r

a
= (*r +-*k

is a well known algorithm for

computing the square

discrete system in Simnon and explore the algorithm"

root of a. Write

2.

Modify the simulation program in exercise 1 so that the stationary solution to

the equation is cornputed and plotted.

3.

The following is a simple Keynesian model of a macro economic system

v (t) =

c(t) * i(t)

+ g(t)

= Y(t-t)
i (t) = b [c(t) -.(t-1)]

c (t)

where y is the gross national product, c consumption, i investment and g

government expenditures. Write the equations as a system of first order
equations and simulate the equations. Investigate the response of the system
to a sudden increase in government spending. What are the influences of the
parameters a and b? Try the values a = 0.75 and b - 0.5, 1 and 2.

4.

Investigate the properties of the following difference equation

x(t+l)

y(t+r)

r
x

t)

yzl

exp [-o

.t(*2 + v2)f

for different values of the pararneter r.

r
, l'

.t

f f,

29

5.

Combination

of

Systems

It is often useful to structure a large problem into smaller subproblems. In

simulation this is done by decomposing a large system into interconnected
subsystems. A subsystem is often represented as a box with inputs and outputs
and the interconnections are represented by directed lines between the boxes.
Such a structure can be represented in Simnon. This is done by adding
declarations of inputs and outputs to the system descriptions. The subsystems
can then be described as a CONTINUOUS SYSTEM or a DISCRETE SYSTEM. A
special type of system is used to describe the interconnections. This system is
called a CONNECTING SYSTEM.

Consider the control system shown

in Fig. 8 which is a

combination

subsystems

discrete system REG

input yref y
output u

END

continuous system

PROC

input u
output y

I
, '

t
?

END

and the system which describes the iriterconnection is given by

.t f',,

ll'l
lt

30

of

two

r
PROC

REG

Figure

8.

Block diagram of an interconnected system'

connecting sYstem CON

"Connecting system for simulation of process
"with PI regulation by system PIREG

PROC

time t
yrIpireg] =1
yIpireg]=yIproc]
uI proc ] =uIpireg

END

Notice that states, variables, and parameters are local variables in each
subsystem. Variables in different subsystems may be specified by adding the
system name in square brackets after the identifier. Also notice that expressions
may be used to describe the interconnections. constructions like

y[reg] = if t<l then 0 else sin(k*v[proc])

are thus possible.
The simulation of an interconnected system is done using the same commands as
was used to simulate difference or differential equations. The only difference is
that it is necessary to activate all subsystems that describe the interconnected
system. This is done bY the command
SYST SYS1 SYSz

CON

The connecting system must be the last system in the list' The order of the
systems is otherwise irrelevant. Contir,tr,'r, and discrete system may be mixed

freely.

+
, lt

AN EXAMPLE

SIMULATION OF

COMPUTER CONTROL SYSTEM

computer control system is conveniently

described as an interconne"t"h'=yJtem. The process may be represented as a
CONTINUOUS SYSTEM and the control computer as a DISCRETE SYSTEM'

A continuous time process ttv

31

discrete system PIREG

"PI regulator with anti-windup
input yr y
output u
state i
new ni
time t
tsamp ts
e=yr-y
v=k* e+i

u=if

v<ulow then ulow else

ni=i+k* erh/ ti+u-v
ts=t+h
k:1
ti: I
h: 0.5

if vcuhigh then v else

uhigh

ulow: -1
uhigh: 1
END

connecting system CON

"Connecting system for simulation of process
"with PI regulation by system PIREG

PROC

time t
yrIpireg]=1
yIpireg]=yIproc l
uIproc]=uIpireg l
END

Listins

5.

Sirnnon descripton of a simple control loop consisting

time process and a discrete time PI regulator.

of a continuous

Listing 5 describes a feedback loop consisting of a continuous time process called

PROC and a digital PI regulator called PIREG" The process is an integrator with
input saturation. The interconnectns ,ar,e described by the connecting system

coN.

4.tf

ln

32

, t'

The following annotated dialogue illustrates how Simnon is used.

Command

Action

syst proc pireg con

store yr y[proc] tpt
simu 0 40
split 2 1
ashow y yr

Activate the system.

Select variables to be stored.
Simulate.

Form two screen windows.

Draw y and yr with automatic scaling in first

window.

Draw upr with automatic scaling in second window.

ashow upr

Notice that the names are local to each subsystem. To distinguish between
variables that occur in different subsystem the name of the subsystem is written
in square brackets as in y[proc]. Variables can be transmitted between subsystem
by declaring them as inputs and outputs.
The results of the simulation are shown by the oscillatory curves in Fig. 9. The
discrete nature of the control actions generated by the computer are clearly

visible in the curves. These curves show that there is a considerable overshoot
due to windup at the integral. This is avoided by telling the regulator what the
process limitations are. The commands

par
par

ulow: -0.1
uhigh: 0.1

changes the necessary parameters. The commands

simu 0 40

area 1 I
show y yr
area 2 L
show upr

,,

shows that the overshoot is reduceflrsignificantly. Compare Fig.

GENERALITIES

9"

."t

Simnon allows three types of s3terl,. descriptions, namely CONTINUOUS SYSTEM,

DISCRETE SYSTEM and CONNECTING SYSTEM. The discretE ANd thc CONtinUOUs

33

T,2
la-

L
!

0"6

+t

f
e
J
o

+)

0
0

10

20.

30.

40

10

20

30

40

0.1
J

o
.;
n

L
{t
c

-0"1
0

Time

Figure

9.

Results of simulation of process control with a pl regulator. The

curves with a large overshoot correspond to an ordinary regulator.
The other set of curves are obtained with a regulator with overshoot
inhibition.

systems may be simulated individually provided that no inputs and outputs are
declared. Interconnected systems may also be described by using the connecting
system. The complete syntax

for the system descriptions is given in Appendix

B.

EXERCISES

Look at the syntax of the comnTands SIST and SIMU using the help command.
what are the differences, dtwen simulation of single systems and
interconnected systems.

The HELP command

has

experimentally.

f
I
34

an hierarchical structure. Explore

y',"

this

3.

Use the command HELP LANGUAGE STRUCT to find the form

system descriptions.

4.

Assume that the variable y is used in two subsystems in an interconnected

system. Construct a simple test example to find out what happens. What
diagnosis is produced? How can the variables be separated?

5.

Consider the system in the example. Repeat the simulation on your own
computer. Investigate the consequences of changing the sampling period.

6.

Study the structure of the system descriptions.

of the different

t
, l)

f,,

35

6.

Advanced Features

Simnonmaybeusedinmanydifferentways.sofarwehavedescribedproblems

whichmaybesolvedbyusingonlyadozencommands.Thisissufficientfor
are however several
demanding problems there
more
For
applications'
many
additionalfeatures.ThesemayallbeexploredbyusingtheHELPcommandorby

readingthemanual.Abriefdescriptionofsomeusefulfeatureswillbegiven
here to indicate some possibilities'

MACROS

Commandsarenormallyreadfromtheterminal.ltis,however,usefultohave

theoptionofreadingasequenceofcommandsfromafileinstead.The
construction
HACRO NAME

Commandl
Command2
Command3
END

thusindicatesthatthecommandsl,2and3arenotexecuteddirectlybutstored
onafile.Thecommandsequenceis"thenactivatedsimplybytypingNAME"

AsanillustrationwegivettrtfoltowingmacrowhichgeneratesFis.9.
t
, l'

macro FIG9
"Generates Fig

'

syst Proc Pireg-con

store Yr Y[Proc] tPr^
simu 0 40/wuP "1, .,
par ulow: -0'1 ll| ^. f '.
Par uhigh:0'1
simu /nowuP

36

split 2 I

ashow y/rurp
show yr y/nowup
ashow upr/wup
show upr/nowup

mark a 2.5 O
mark "Time t

markall
mark v "Control variable u
mark a t 7.5
mark v "output y and yref
END

Notice that lower case letters may be used to get a more readable code, although
Simnon does not distinguish them from upper case letters. Macros

all figures for this report are given in Appendix

Macros are useful

of a dialogue.

for

for

generating

D.

documentation. They are also convenient

for

simplification

Command sequences, that are commonly used, may be defined as

macros. A simple macro call will then activate a whole sequence of commands.
Macros may also be used to generate new commands.

The usefulness of macros may be extended considerably by introducing

commands to control the program flow in a macro, facilities for handling local and
global variables and by allowing macros to have arguments. By having commands

reading the keyboard and for writing on the terminal

implement menu driven dialogues using macros.

for

Even a casual user

is strongly

recommended

it is

also possible to

to learn simple uses of the macro

facility.

INTEGRATIONALGORITHMS f
, lt

Differential equations are solved in Simnon using numeric integration routines. A

predictor corector formula by Hamming with autonatic step length adjustment is
normally used. The initial value of the' step length is chosen as one hundredth of
the integration interval. A differfrt firitial step size may be chosen by an optional
argument

in the

ln '|
command SIMU1 lh tde algorithm the step length

is reduced until

the difference between the prediction and the correction is sufficiently small. The
tolerance may be set by the command ERROd.
37

It is also possible to choose other integration algorithms by the

ALGoR {r{AHPc

RK

command

RKFrx I DAS}

where HAMPC is Hammings predictor corrector method, RK is a fourth order

Runge-Kutta algorithm with automatic steplength adjustment. RKFIX is also a
fourth order Runge-Kutta algorithm but it has a fixed step length. DAS is an
algorithm for integration of stiff equations, i.e. differential equations with both
slow and fast modes. Further details on the different algorithms are given in the
Simnon manual.

FORTRAN SYSTEMS

The Simnon language is simple, easy to use and reliable because of all the
diagnostics that is built into it. The language has however a limited expression
power. There are no possibilities to control program flow, there are no arrays
or other data types and there are no procedures. For models whose descriptions
require a more powerful language it is possible to interface Fortran routines to
Simnon. This also makes it possible to use library subroutines like Eispac and
Linpac in the simulations. The Simnon manual describes in detail how to do this"

STANDARD SYSTEMS

There are several standard systems in Simnon. A list of the available systems is
normally displayed on the screen when.*the system is started. A few of the
systems are listed below.
DELAY
FUNCl
LOGGER

NOISEl
OPTA
IFILE

\
' '
- time delay
- nonlinearity defint Uf " t"ut"
- logging of stored variables
- white gaussian noise generator
- optimizer
- discrete system which reads output variables from

file

There are also more sontrisiag systems tike linear quadratic

gaussian

regulators, self-tuning regulatos in some implementations. A description of the

standard systems is given in the manual. The command HELP SYSTEMS also gives

38

information about the systems. The standard systems are often implemented
Fortran systems.

as

GLOBAL VARIABLES

All variables in

Simnon systems are local. When using Fortran systems or

standard systems it may, however, be necessary to transfer global data. Global
variables are used for this purpose. The global variables are set by a LET
command. An example illustrates the use of global variables and standard
systems.
Example 6.1

Use of standard systems

Assume that we want to include a function defined by a table in a simulation. This

is conveniently done by the function FUNCI. This function has one input u

and

y. The function has one global parameter N.FUNCI which gives the
number of entries in the table. The arguments are specified by the local
parameters uit, ui?, ... . The corresponding function values are specified by the
one output

arguments gil, giZ,... . The local parameter ORDER specifies staircase (order = 0)
or linear (order = 1) interpolation.
The following dialogue illustrates how the function may be used.

Iet n.funcl = 4
syst FUNC1 FUNCPLOT
par uil: -3
par gi1: -1
par ui2: -1
par gi2: -2.8
par ui3: I
par gi3: 1.5
par ui4: 4
par gi4: 2
par order: 1

\
, l'

The global parameter n.funcl must, be assigned a value before the system is
activated by the SYST command. The local parameters may be changed by the
PAR command as parameters in f4inary Simnon systems.

,
frn
Itt

39

f (x)
2

-1

-2

-3
0

-4

2
x

Fisure 10. Graph of a function defined by a table generated using the Simnon
function FUNCI.

with
connecting system

time

FUNCPLOT

uIFUNCl]=x
yp=yIFUNCl]
END

the commands

axesh-44v-33

plot

yp

simu

-4

, '

t
I

then generates a graph of the function, see Fig. 10'

,11, 1,.

40

ORGANIZATION OF LARGE SIMULATIONS

There are several facilities which are useful when working with large systems or
with large amounts of data. System descriptions and macros are stored as text

files. These files have the file extension .T on the VAX/VMS system. The
subsystems also have a name which is the identifier given on the first line of the
system description. See Appendix B. Notice that the file name and the system
name may be different. This is very useful when simulating different versions of
a model because the same macros and the same connecting systems may be used.
The selection of a particular model is done when the systems are activated by the
command SYST. An illustration is given in the macro FIG9 which is listed in the
appendix. A PI regulator with the name REG is stored in a file called PIREG and a
system with the name PROC is stored in a file called INTEGR. The systems are
activated by the command
SYST INTEGR PIREG

CON

The variables of the system are labeled

makes

it

by

[REG] and [PROC] respectively. This

possible to use a standard connecting system for different processes and

regulators.

SAVING AND RETRIEVING PARAMETERS AND INITIAL VALUES

The command SAVE stores parameters and initial values in a file. The values may

be retrieved by the command GET. These commands are very useful when
working with large models, because parameters and initial values are not
introduced manually. Assume for example,that the systems FUNC1 and FUNCPLOT
have been activated as in bcample 6.1 by the command
SYST FUNcI

FUNcPLoT ,t 1

'

The command
SAVE FUNCPAR

then generates a file FUNCPAR.T,,pf .tthe form

(

t',

4L

IFUNC1]

UIl: -3.
UL2: -1.
UI3: 1.

UI4:4.
GI1:
GT.2:

-1.
-2.8

GI3: 1.5
GIA:2.
order: 1.
IFUNcPLor]

It is convenient to edit files of this type for

parameters and initial values when

simulating large systems.

DOCUMENTATION

It is often useful to keep a running log of an interactive session. The command

SY{ITCH LOG ON

generates a file of all commands in a session.

The scale factors in the graphs are computed by an algorithm. The same axes are
obtained in horizontal or vertical direction if the horizontal range is divisible by

four and the vertical by 3. It is possible to get scale factors 1, 2 and 5 only by
the command
TURN

5125

ON

It is possible to add text to the axes by the command MARK. This is used in

the

macros which generate the curves in the report.

It is also useful to have access to operating system

commands under Simnon. In

the Vax implementation this is done]gimply' bV typing $ followed by any operating

system command.

t f,

t',

42

7.

lmplementation

When using Simnon it is helpful to have some insight into how the program
works and how it is implemented. A brief description is given in this chapter.

HOW

SIMULATION IS ORGANIZED

When the command SYST is given the equations describing the system to be
simulated are first sorted so that all calculations are done in the correct order. It
is possible to store the sorted equations in a text file by using the optional
argument [/<filename>]. The command
SYST INTEGR PIREG CON/DUHHY

generates a file called DUMMY with the content:

SORTED

INITIAL

SORTED

TIME-INDEP. EQUATIONS

cON

EQUATIONS

yr[REG] =

SORTED DERIVATIVE EQUATIONS

PROC

coN

REG

coN
PROC

Y=x
y[REc] = y[PRoC]
1+
e=yr-y
v = k*e + i , l, t
u = IF v(ulow TllEll ulow EI-SE IF v<uhigh

u[PRoc] = uIREG]
pr = IF u<-0.1 THEN -0.1 FrsE IF u<0.1
dx = upr

THEN

THEN

u EISE 0.1

ELSE

uhigh

SORTED CONTINUOUS EQUAT,XONS

ln

'

SORTED DISCRETE EQUATIONS

REG

ni=i+k*erh/ti+u-v
ts=t+h
43

The subsystem name is given to the left and the equation to the right. The
equations are sorted so that the calculations may be performed in sequential
order. The equations .are then brought to the standard form of a system of first

order differential equations, which are then integrated using the chosen
integration routine. The state variables of discrete time systems are constant
between the sampling instants. They may change discontinuously at the sampling

instants. The integration is therefore carried out only to the nearest sampling
instant. The states of the discrete systems are then changed ahd the integration is
continued.

ALGEBRAIC LOOPS

When systems are interconnected it may happen that the equations can not be
sorted so that the variables may be obtained by sequential computations. A simple
example, which illustrates what may happen is given below.

continuous system Sl

Input

Output y

V=u
END

continuous system 52

Input

Output y

V=u
END

connecting system

yr=1
uI S1 ]=y.+k*y[s2]

uI 32 I =
k

0 .5

v[sl]

END

, l'

When the command

SYST

51 52

is given the error

sort the equations result in

44

'4,

diagnosis "{Cbhqic loop detected"

is given. The attempt to

i,*-In it io lize

Reod commond

Normol

Mocro

Keyboord

Doto bose

Decode commond

Commond SYST

Commond ST0P
oaa

Fisure 11. Skeleton flow chart for Simnon

51
c
SZ
C
51

Y=u

u[S2] = u[Sl]

Y=u

u[S1] = yr * k*y[S2]

Y=u
The variable u[Sl] thus can not be solved sequentially. A simple calculation
shows that the variable is given by the following algebraic equation:

u[S1]=]F*k*u[Sl]
This explains the name algebraic loop. In this particular case it is easy to solve
the equation if k # 1. In the general case the algebraic equation may be nonlinear.
It is then difficult both to determine if a solution exist and to find it if it does.
Simnon will not attempt to find a solution. It will just give an eror message. The
presence of an algebraic loop is often an indication of poor modeling. Models with
algebraic loops may be integrated with the qroutine DAS.

t
, l'

HOW THE PROGRAM IS ORGANIZED

for

is shown in Fig. 11. The main loop reads a

command, decodes it and nerforry the required actions. All parts of Fig. 7 except
the action routines are impler$ntdd,, as a package of subroutines called Intrac.
These subroutines perform command decoding, file handling, listing and plotting.
Intrac also contains the macro facility. A supmary of the commands in Intrac and
schematic flow chart

Simnon

45

Simnon is given

in ApPendix

E.

7000 lines of Fortran code, which compiles to

about 90 kbytes of code. Simnon itself has a source code of about 25 000 lines of

The source code

for lntrac is about

Fortran code, which compiles into 360 kbytes of

t
, l'

t',

46

,n

code.

8.

References

Simnon was developed

control systems.

in connection with a project for computer aided design of

See

,strm, K J (1983):
Computer aided modeling, analysis and design of control systems
perspective. Control Systems Magazine, May, pp 4-16.

Simnon inherited many ideas from the tradition of analog and digital simulation in
the control engineering field. When the work on Simnon was started there were

a number of digital simulation languages like CSSL and CSMP available. These are
described in the book
Korn, G" A. (1978):
Digital Continuous Systern Simulation. Prentice Hall. Englewood Cliffs,
New Jersey

which contains much useful information on simulation. Languages like CSMP and
CSSL were, however, implemented using batch calculations. Since the languages
largely were inspired by analog simulation techniques they also inherited several
constraints imposed by the analog hardware. For example the state space notation
which is so natural was not supported 6y p.op"r language constructs. When
Simnon was developed the main ida was, to provide a tool which stpports the
state space notation with good,rnBh-rffachine interaction based on interactive
computing.

first version of Simnon was defiped and implemented by Hilding Elmqvist as

an MS dissertation in 1972. The,wgrk was continued by Elmqvist and a usable
The

language and a manual were av{lbfe',in 1975:

Elmqvist, H (1975):
SIMNON, an interactive simulation pfogra

for nonlinear

systems.
47

Report TFRT-3091, Dept of Automatic Control, Lund Institute of

Technology, Lund, Sweden.

A good description of Simnon is also found in

Elmqvist, H (19771:
SIMNON - An interactive simulation program for nonlinear systems
Simulation '77, Montreux, Switzerland, June 1977.

A new Simnon manual by Elmqvist is in preparation.

in teaching at the department of

Automatic Control at Lund Institute of Technology ever since. It is increasingly
The Simnon language has been used extensively
being used at other schools and industries.

An innovative use of Simnon is given in the textbook

.A,strm, K J and Wittenmark, B (1984): Computer Control Theory

All graphs for simulation results in the book are

generated

by Simnon

Macros

which are accessible to the students. They can then easily change parameters and
modify graphs.

The interaction principles used in Simnon, which are based on commands

macros were developed in a more general CAD context. See

and

Wieslander, J (1979):
Interaction in computer aided analysis and design of control systems.
PhD thesis, Report TFRT-1019, Dept of Automatic Control, Lund
Institute of Technology, Lund, Sweden.
and

Wieslander, J (1980a):

Interactive programs - Genral guide. Report TFRT-3156, Dept of

Automatic Control, Lund Instifrrte gf Technology, Lund, Sweden.

'l

The program Intrac, which is the core of the interaction with the user,
described in
Wieslander, J and ElmqvisUfH 11978):
INTRAC, A communicatiorf pddrle for interactive programs. Language
manual. Report TFRT-3149, Dept of Automatic Control, Lund Institute
of Technology, Lund, Sweden.

48

is

APPENDIX A

Syntax for Simnon Commands

A list of the commands, their syntax and brief descriptions are given below.
Some Intrac commands (denoted bV t) are also included in the list. More details
about the commands are obtained using the command HELP.

are listed in Appendix D. The list is valid for

All Intrac

implementations

commands

on

VAX-11

systems. The following notations are used:

{op1

1... loptt}

Defines different alternatives of which one must be given.

t....1

Parts within squared brackets are optional and can

A star indicates that the previous part can be repeated"

}*

be

omitted.

Defines arguments

for the commands.

, l,

l^,

f,

49

ALGoR {HAMPC IRK lRKFrx IDAS}

To select integration routine.
HAMPC - Hamming predictor corrector (default)
RK - Runge-Kutta variable step size
RKFIX - Runge-Kutta fixed step size
DAS - Integration routine for stiff systems
HAMPC

ALGOR

RKFIX
DAS

AREA <row> (column)

To define the plotting area to be used next, see SPLIT
1
1

AREA

2
3
Column

Row

ASHOW {[<start><stop>] {<variable>}* [(<variable>)l [-MARK]l-LIST]

[/<ritename>]
To plot stored variables with automatic scaling. Similar to SHOW
Start

Var

ASHOW

Lt

50

sr

MARK

Varidble

Stop

Filename

AXES [<axis spec> [<axis spec>]l

To draw axes.
<axis spec>"= {HlV} <min value> <max value>
H - horizontal, V - vertical
Min value
Min value

Max volue

Max value

DISP [({DISITPILP}[{FFILF}])l [{<variable>}*]

To display variables.
DIS - display (default)

TP
LP
FF
LF

teleprinter
line Printer
form feed (default when no variables are specified)
line feed (default when variables are specified)
If no variables are given all variables are displayed.
DIS

FF

rP

LF

Variable

LP
DISP

I
l'

t',

f,

51

EDIT <filename>

To edit a file. The editor has two modes, INPUT and EDIT. The mode is
changed by entering an empty line. In the edit mode it is possible to change
the current file. The following commands are available:
Append string to current line
AIPPEND] <string>
Line pointer to bottom
Blorrl
Replace string x by string y
clHANGl l"lvl
Delete n lines
D[EL] <integer>
Drs [oNloFF]
EIxrr]
F[lND] <string>
I[NS] <string>
L[OC] <string>
LEAVE
N[EXT] <integer>
OIVERL] <integer>
PIRINT] <integer>
RIETYP] <string>

rloPl
EDTT

ERROR

Echo check enable/disable

Leave editor with updated file
Find string at the beginning of a line
Insert a line below current line
Locate string anywhere in a line
Leave the editor without uPdate
Move line pointer down n lines
Overlay n lines by new text
Print n lines on display
Retype current line
Move line pointer to the top of the file

File name

<error bound>

To choose
ERROR

eror

bound

for integration routine. Default value is 0.001.

Error bound

GET <filename>

To get parameter values and initial values from a file that previously
been stored using SAVE
GET

or

edited.

File name

, l)

i
|'

52

.,t

'

'.

has

HcoPY {[<MoDE>] | [<factor>]

MoDE::=(Q
'iwirCn' lLlRlLP)

<swlTcH>]

rsHow)
is scaled with
on display' The hardcoPY
Make a hardcopy of curves interval
(0.5, 1.6) The cmment is obtained
<factor), *nillil"v be in the
on the hardcoPY too'
= 1b

Factor

HCOPY
ON

OFF

sH0w

, l,

t f,,

53

HELP

[DEv] IFEED] [PRoMrINSLASK]l)l IKEY1 IKEY2..l..l

DEV::={DISILPITP}
FEED::={LFIFF}
PROMT::={ASKICONT}
*6y1';={SIMNON COMMAND IINTRACISYSTEM}
KEY2::={DETAIL OF SIMNNONINTRAC COMMAND}
[(

To get more information about the commands, the editor, Intrac, Simnon, and
the standard systems. A menu of commands is obtained by typing HELP.
Nslosk

Dts
LF

ASK

FF

fP

o
HEL

Detail of simnon

Intrac command

Simnon command
INTRAC
SYSTEM

INIT <state variable>: {<number>l<variable>}

To change the initial value of a state rariable.
T

I
INIT

.Number

State voriable
Varioble

i
54

/,,

LET {<variable>=}* {<number> [<operator> <number>]

l{+l-} <number>l <identifier> [+<integer>]
lcdelimiter> lcunassigned variable>)
<operator>::= {* l-l- l/l ^}
An Intrac command that also is used to set global parameters
standard systems.

@
LET

for

the

Number

Variab le

Number

nlege

ldent if ier

Delimiter

Unassigned varioble

LISr [({DISITPILP}[{FFILF}])l {<filename>}*

To list textfiles. The first arguments are the same as for the command DISP.
When typing the command LIST the files are sent to a print file. Type the
command LP to initiate the printing. ,*

Dts

LF

rP

FF

LP

Filename

Ltsr

55

LP

Initiate listing on lineprinter

DISP, LIST, LOG

or

of the print file generated by the

commands

PRINT.

MARK

To introduce text into a plot. For syntax type HELP INTRAC MARK.
MARK

NEWS

To obtain news about Simnon.

NEWS

PAR <parameter>

: {<number>lcvariable>}

To change a parameter value.

Number
PAR

Porameter
Var

[ {<variable>} *

ia ble

[(<variable>)] l
To select variables to be plott when making the command SIMU:
Examples: PLOT XL X2 give5 Xl nd X2 as functions of time while PLOT
X1(X2) gives Xl as function of X2.

PLOT

Vorioble
Varable
PLOT

56

PRINr [({DISITPILP} [{FFILF}])l <filename> [<lines>l

[/<start time>]
To list file generated with the commands STORE+SIMU. <lines> lines starting
from (start time> will be printed. The other parameters are the same as for
DISP.
Dts

LF

TP

FF

LP

@
F

PRINT

ilenome

Stort time

Lines

SAVE <filename> [<systemname>] [-{PARllNtT}]

To save parameter values and initial values for a given

system
GET'
command
the
by
used
be
<filename>
to
<systemname> on a file named
INIT.
or
PAR
option
the
with
are
saved
only parameters or initial values
PAR

System identif ier

lN

lr

Filenome

\
, t'

t
l*,

f,"

57

SHOW {[<start><stop>] {<variable>}* [(<variable>)] [-MARK]l-LIST]

[/<filename>]

To plot stored variables from file <filename>. To be used with the command
STORE. The specified variables are plotted frorn <start> to <stop> time. If
MARK is used the different variables are numbered on the plot. The
LlST-option lists the names of all variables.

Voriable

Stop

@
Variable

@
LI

ST

Filename

MARK

@
@

\
l'

hI

58

t,

SIMU [<start time> <stop time> [<time increment>]]

[-{coNTMARK}[CONTIMARKI [/<filename>] [<store increment>]
To simulate the system from <start time> to <stop time> using the maximum
stepsize <increment> (default (stop time - start time)/100). Using MARK the
variables defined by PLOT are numbered. With CONT the simulation is
continued with the previously obtained state variables as initial values.
When specifying <filename> then the plotted variables are stored in
<filename> with <increment> as sampling interval.
Time increment

Stort time

Stop time

SIMU
CONT

coNr

Filename

Store increment

SPLIT <rows> <columns>

To split the screen into maximum six plotting areas

<rows>:={1 | 2l 3} default
<colums>:={112} default 1

Row

Column

tl

f.r
^,r

59

STATE {{<variable>}* [(<variable>)] <option>ll

-CHECK <time increment> I-STATUS)
<option>: :={SLOW IFAST I SOLVE}

A special command, which is only used with the integration routine DAS.

FAST

Varid ble

Voridble

Time increment

STATUS

STOP I
To leave Simnon.
sfoP

SToRE [{<variable>}* [-ADD]l

To select variables to be stored at each simulation. With ADD new variables
can be added to a previously defined list of variables. The variables can be
displayed using ASHOW or SHOW and printed using PRINT.
ADD

,l

STORE

I (,.t

l',

60

swrrcH {cLocKlDArElEcHolEXEClLoGlrRAcE} {oNoFF}

To control the execution of Intrac.
CLOCK - Adds time to hardcupy output. Default OFF.
DATE - Adds date to hardcopy output. Default OFF.
ECHO - Macro commands are echoed. Default OFF.

EXEC -

Commands executed while typed during macro generation.

LOG TRACE -

Default OFF.
Executed commands are logged on line printer. Default OFF
Affects ECHO and LOG.
CLOCK

DATE
ON

ECHO
SWITCH

EXEC
OFF
LOG

TRACE

SYST {<identifier>}* [ -<option>l [/<filename>l

(option):={EDIT I EXIT}

To define the system. The subsystems are compiled. If there are several
subsystems the last one has to be a connecting system. EDIT means that the
compiler goes into the editor for each file. If filename is specified the sorted
equations are written into a text file.
ED

EXIT
SYST

I d ent

ilename

if ier

TEXT '<any string not contai.,irrs,s$gle qdote>'

To include text on paper plot.

TEXT

Any string not contgining singte quote

4

ln,

61

ruRN {DARKlDrslNocoDElovFLO|PLCOMlsl25lrrMrNG} {oNloFF}

To turn on and off switches.
5125 Selects a limited set of scale factors (1., 2.,5.1when
choosing scales on exes. Default OFF.

DARK

ON means that plotted curves will not be connected

sampling instants. Default OFF.

DIS

lnforms

between

if the user has a graphic display. Default ON.

OVFLO ON means that the simulation stops if overflow occurs in the
SIMNON

calculations. Default

OFF"

DARK

Dt5

ON
rURN

OVFLO
OFF
PLCOM

s 125
TIMING

t
, l'

l^,

62

f,,

I
I

the

APPENDIX B

Syntax of System Descriptions

Simnon allows three types of systems: CONTINUOUS SYSTEM, DISCRETE SYSTEM

and CONNECTING SYSTEM. The following Bachus-Naur notations are used to
describe the syntax.
syntactic unit
denotes
I

{
*

exclusive or
optional element
compulsory element
repetition

The following syntactic elements are needed to describe the systems.

LETTERS

<Ietter>:: =AlB c
alb c

<digit>::=

F G H I J
o
j
d e
h
011 2 3 4 5 6 7 8
D E

KlLl ll I N lo I plolRls lr lu I vlx I Yl z

lk I I lm ln I o lp lq l'l " I t lu lv I x ly I z
I

IDENTIFIERS

<identifier>: : =(letter) | <identifier><letter>

<identifier><digit>,F
1h

VARIABLES

, ,t

<system

't

'

identif ier): : =(identifier>

<simple variable>: : =(identifier>

<variable>: : =<simple variable> | <simple variable>[system

identifier>]

t.t (

l',

63

The generic form of the system descriptions are:

SYSTEI{ <system identifier>

<simple
INPUT
variable>* l
t
<simple
vari abl e>* l
OUTPUT
t
<simple
abI e>* l
STATE
vari
t
<simple
abl e>* l
vari
DER
t
<simple
e> ]
TIME
variabl
t

CONTINUOUS

fturrtrl

of initial values for state variablesl

ls8il"."tton
t Computation of auxiliary variables]
t Computation of output variables]
t Computation of derivatives]

Parameter assignment
IInitial value assignment]
END

DISCRETE SYSTEH

<system

identifier>

IINPUT <simple variable* ]

[OUTPUT <simple variable>* l
ISTATE <simple variable>* l
INEfl <simple variable>* l
ITIHE <simple variable>]
TSAHP <simple variable>

"denotes the new values of

the states
"denotes the next sampling

instant

initial values for state variables]

initial values for output variables]
initial values for the TSAHP-variable]

t Computation of auxiliary variables]

t Computation of output variabl ef]
fComputation of new values of ttrg slatesJ
Updating of the TSAHP-variabl
IHodification of states in continuous subsystems]
Parameter assignment
IInitial value assignment]
END

tl r
64

CONNECTING SYSTET{

<system

identifier>

t TIHE <simple variable>]

t Computation of auxiliary variables]
t Computation of input variables]
t Parameter ass i gnments]

END

Note that the order of the computations and assignment is unimportant because
the equations will be sorted automatically. The section Initial makes it possible to
compute initial values to state outputs and TSAMP. These variables can normally
not be assigned an expression.

It

r',

65

APPENDIX

Standard Systems

This appendix describes the standard systems DELAY, FUNCI, IFILE, LOGGER,
NOISEI, OPTA, and STIME. These systems are written in Fortran and linked into
Simnon.

DELAY

This discrete time system simulates pure time delays. Old values of the signal to
be delayed are stored in a vector. Delayed values are then generated by
interpolation. The system admits two interpolation schemes due to Hermite and
Aitken. In Aitken's scheme a Lagrange polynomial is fitted to the stored values. In
Herrnites method the values of the derivatives are stored together with the
function values. The delayed values are then determined using an interpolation
polynomial, which agrees with the function and its derivative at the stored points.
Hermite's method requires that derivatives of the function are also stored.
The function admits delay of many signals. It has the following global variables.

nl.delay

Number of variables using Hermit interpolation

n2.delay

Number of variableS ..rring'Aitken interpolation

'tlie
a[ocation area to be used for saving
Number of elemetj t"
old values. Try to use as many as possible, SIMNON gives an
error message if too much space is used.

space.delay

These variables must be assigned V

activated.

66

t^r

a LET command before the system is

The system has the following local variables

INPUT:

u1,u2,...
du1du2,..
td1,td2,...

Variables to be delayed (n1+n2)

Derivatives of the variables (n1)
Delay times (see below) (nl+n2)

OUTPUT:

yL,y2...

The delayed variables (n1+n2)

The outputs from the system DELAY are all zero

for

t<td, but other values can

easily be set in the equations (see example below). The following

example

illustrates how the system DELAY can be used.

Example: Assume that the following systems have been defined

continuous system

SYSI

END

connecting system

time t
td1[detaY]=t-5

u1

CONN

[de I aY]=s in ( t )

u[sysl] = if t<5 then 1 else yl[delay]

END

The global variables

for the system delay are

assigned by the commands

let nl.delay=g
Iet n2.delay=1
Iet space.delay={5g
The systems are activated by the command

syst sysl delay conn

The signal u[sysl] is then given by*

u[sys1] (t)

lt

I
=

ift<5

sin(t-5)
,r'^

ift>5

67

FUNCl

This is a continuous time function which makes it easy to introduce functions in

tabular form. The system has one global variable:
n.

funct

which gives the number of tabulated function values. The local variables are:
INPUT:

Argument value

u
OUTPUT:

Function value

v
PAR:

uil, ui2, ...

git, gi2, ...

order

Table of argument values

Table of function values
Order of interpolation (0

or 1)

The following example illustrates how the function is used

Example: Assume that the following systems are defined.

continuous system

SYS1

END

connecting system

time t

u t func 1 ]=
u t sys 1 ]=yI func I

CONN

END

The global variable of the system FUNC1 is assigned by the command

let n.funcl=S
The systems are activated by the

"o.*"nif

syst sysl funcl conn

,
, t' '
The parameters of the system FUNCI are defined by the commands

68

rt

par
par
par
par
par
par
par
par
par
par
par

ui1:0

gil:0

ui2: 1
gi2: I
ui3: 2
gi3: 4
ui4: 3
gi4: 9
ui5: 4
gi5:16
order: 1

"Use

linear interpolation

The parameters of the system FUNC1 are saved on file TAB1.T by the command

save

tabl funcl

Another illustration is given in Example 6.1.

IFILE

This is a discrete time system which reads variables stored in a datafile. It is

useful for example when comparing simulation models with real data. The system
has two global variables:
Nurnber of columns to be read from
number of output variables
File name for input data file

n.ifile
fname"ifile

file or equivalently the

The local variables are

OUTPUT
c1

c2

PAR:
dt 1
dt

lst colu4n in the file

2nd column in the file
Time for first input relahve to the start time (default: O.O) '
Distance between inputs {deffult: 1.0)

TSAMP:

ts

Time

for next input

The actual value of n.ifile is fixed during the execution of the SYST command. The
.'
value of fname.ifile can be charlged
't 'r etween simulations. The simulation terminates
if the input file is exhausted.

.t

69

Several systems with different system identifiers can be active simultaneously.

The global variables should then contain actual system identifiers. The following
example illustrates the use of IFILE.

is stored in the file FDAT.D, which has 10 columns

and that we want to use the third column as input to a simulation. Let the
Example. Assume that data

following systems be defined.

continuous system

SYS1

END

connecting system

time t
u[sYs1]=s3[ifile]

CONN

END

The global variables of the system IFILE are assigned by the comrnand

let
Iet

n. ifile=10
fname. ifile=fdat

and the systems are activated by the command

syst sysl ifile

conn

The file used as input to IFILE can be created in Sirnnon using a store command.
It can also be generated in the system identification package ldpac.

LOGGER

,F

This is a discrete time system which can be used to sample d

"t
X
Simnon variables on a file.
'
, t' t

"".t"

arbitrary

The system has one global variable:

file.npoint
which gives the number of dat{ qoints stored in the current STORE file with
Jt; t '
by LOGGER after each simulation. It can be
name <file>. The global variableT is set
displayed by the WRITE command. (This value rnust be known if the generated file

70

is to be used by ldpac.)
The local variables are:
PAR:

Time for first sampling relative to the start time

(default: 0.0)
Sampling interval. If dt=0.0, no fixed sampling is done,
and all points are saved (default: 0.0)

dt1
dt
TSAMP:
ts

Time

for next sampling

When it is desired to store data the system LOGGER should be included as a

system in the SYST command. Since it has neither INPUTs nor OUTPUTs, it need
not be connected in the CONNECTING SYSTEM. A connecting system must,
nevertheless, be present even if only one system is compiled together with
LOGGER. The connecting system can then be empty (3 lines). The following
example illustrates how the system LOGGER can be used.

is simulated, which has a variable x and

that it is desired to store the values of x every 10th second starting at time
| = 2.O. Assume that the following systems are defined.
Example. Assume that

a system

continuous system

SYS1

SYS1

END

connecting system

time t

CONN

END

The systems are activated by the cornmand

syst sysl logger

conn

The parameters of the system LO clfn -ar set by the commands

par dt[Iogger]:10
par dt1[loggerf:2
The command request the variabp x of SYS1 to be stored.

store x[sys1]

t',

The following simulation command executes the simulation with the option that the

71

variable is stored in FIL1

simuO1500/fifl
Notice that the number

of stored variables are available as the global

variable

fill.npoint. The value of this variable is found as follows:

write fill.npoint
149

NOISEl

This is a discrete time system, which generates a sequence of independent

random vectors" The components of the vector have either a normal or
rectangular distribution. The system has the global variables
Number of outputs

n.noisel
nodd.noisel

Initial value for the generator (should be an odd,

positive integer)

The local variables are

OUTPUT:

e7, e2, .,.

PAR:

stdevl, stedv2,

The noise vector

.

Standard deviations

for the outputs (defautt:

1.0)

for first output relative to start time (default:

dtl

Time

dt

Distance in time between outputs (default: 1.0)

same

Reset switch. When same > 0.5, the noise generator

is reset (default: 1.0)

rect

Type switch. If rect t0.5, rectangular noise in

the interval (0, stdev) is generated instead of
white noise (default: 0.0)

TSAMP
ts

0.0)

Time

Note that the switches

for net otrtptrt

g-

and

g!

influence the entire noise vector. Control of

individual components is not possible.

The global variable nodd.noise|'1r/.na"ted at the end of each simulation. The

value of nodd.noisel can be changed between the simulations. The following

72

example illustrates how to use the noise generator.

Example. Generate two white noise input signals to the system SYSI. The signals

should start at

t = 0.5 and change with a sampling period of 3 time units. Assume

that the following systems are available.

continuous system

SYSI

END

connecting system

CONN

time t
u1 t sys 1 l =e1 t noisel l
u2 t sys 1 l =e2 t noisel l

END

The global variables of the system NOISE1 are assigned by the commands

let n.noisel=2
Iet nodd.noisel=25831
The systems are activated by the command

syst noisel sysl

conn

and the parameters dt and

dtl of NOISE1 are set by

par dt:3

Iet dt1: 0. 5

STIME

This system stores clocktime and cpu-ti."'r, the two Simnon variables MSCLOCK
and MSCPU. lt is useful for tlming rof simulations and for investigating
computational efficiency. No conrr""/ig !yrt"* is needed. It is sufficient to include
STIME in the SYST command" The variables MSCLOCK and MSCPU are set to zero
at the start of each simulation.

I
I

73

OPTA

for the problem of minimizing the function

This discrete time system is a tool

J(p)

subject to the constraints

st(n) f o

i = 1, ...,

lrl.

The system OPTA performs the minimization recursively. The system has old
values of the parameters as states. It accepts values of J and g as inputs and it
generates new values of the parameters, which will give a smaller loss function
as outputs.

The global variables are:

Number of parameters (max 10)
Number of constraints (max 10)

npar.opta
ncons.opta

The major local variables are:

INPUT:

The value of the function to be optimized

The values of the constraints

loss
con1,con2
OUTPUT:

The new values of the parameters

This variable is set to the sampling time

pr,p2
tbeg
PAR:

Sampling period

tinc

There are also several additional

parameters,

which will influence the

optimization. They are listed below.

XMl XM2 .".

LAM1 LAM2
DFN

(1 1 ..) Scaling factors, see HH and

...

(0 0 ..) Lagrange muftipers, initial values

(-0.5) Controls initial st+ at thB first linear minimization; ,should give
an estimate of the likely edtrction in function value, Af; there are two
possibilities:
DFN<O DFN itself is an estimate of Af
DFN<O ABS(DFN).f is taken as an estimate

TINC

HH

EPS

74

EPS

of Af

(1) Length of sampling interval of OPTA

,t
't
(0.005) Step length fpf alculation
of the gradient by differences; the
step length for each component Pi is XMI.HH
(0.01) Stopping criterion

for

unc,onstrained minimization

- is

satisfied

when the change in each component Pi is less than XMi.EPS

PRIN

(1) Controls printout on line printer; every ABS(PRIN):Ih iteration is

printed; if PRIN<O only function values are printed; if PRIN>O also P
and the gradient are printed; if PRIN=O there is no printout

EVMAX

(10000) Maximum number of function evaluations

CEQ

(0) Number of equality constraints

(1) Constant used

problems

DELTA

in the modified function; only used for

(0.01) Stopping criterion

is less than DELTA

constrained

for constrained minimization - satisfied

when

TEST
RESET

DARK

(1) The states are reset to their initial values

if RESET>O

(1) There is no trace on the display at the sampling points of OPTA if

DARK>O

MODE

(1) Controls the initialization of the approximation of the

second

derivative, H

MODE=I
MODE=3
LPLOT

H is set equal to the identity matrix initially

the H-matrix from the previous minimization is used

There is no plotting on the display when the optimization routine

calculating derivatives by differentiation if lplot>O.0

For a detailed discussion of these parameters and their use we refer to

is

Glad

(1e74).

The system OPTA can be used in many different ways. Typical applications are to

adjust regulator parameters for optimized performance and to adjust model

parameters in connection with model fitting. In such cases it is necessary to
perform a simulation in order to obtain the values of lossfunctions and
constraints.

\,

A starting value of the parameter yctqr p is given to initialize the optimization.

The system is then simulated for this parameter value. The criterion and the
constraints are evaluated. The optimization routine then uses the value of the
loss function J and the constraint g to compute a new value of the parameter
vector. The process is then repqat7d with the new parameter. The procedure is
illustrated by Fig. C-1. The critfioh'J and the constraint vector g are connected
from the system to the optimizer while the output of the optimizer is the
parameter vector p, which is an input to te system. At each sampling point of
75

System

0ptimizer

Figure C-l

the optimizer it uses the current values of J and I to compute the next value of
p. If the criterion or constraints depend explicitly on the parameter vector p, a
vector pd, containing delayed values of p, is used. The system can consist of a
nurnber of subsystems. The connections between these subsystems, as well as
connections needed to forrn J and g frorn output of the subsysterns, are made in
the connecting sYstem.

An example illustrates how OPTA is

used.

a control system, whose block diagram is shown in Fig. C-2.

Let the purpose of the controller be to keep x, as small as possible despite the
disturbance v, which is an impulse disturbance. This is equivalent to setting
xr(o)=1. The criterion is
Example. Consider

J= I

2
2

dr

It is also assumed that the total control effort is limited

s=[0

<
dt - u-.
llm - 0

T)

u-.
Ilm = 0.5

The gains KO and KO should be chosen to satisfy these demands. The Simnon
prograrns required to solve the problem are given below.

.tl

76

x1

Figure C-2

continuous system

state xl x2 z w
der dxl dx2 dz dw
input kd kp

IHP

output y
output y=1!
Y=x2

dlmamics

u=-kd*x1-kp*x2
dx1=-x1+u

dx2=xl
dz=u*u
dw=x2*x2
END

connecting system
time tim
wt: 1O
I oss

CONN

Iopta ]=wtr,wI imp]

ulim: .5

conl I opta]=z I imp]-uI im

kdI imp]=p1[opta]
kdI imp]=p2[opta]

lt

, l,

t=tim-tbeg[opta]

END

,,

To do the optirnization the sloaf (rariables of OPTA are first assigned by the
commands

77

let npar.opta=2
Iet ncons.opta=l
The systems are then activated by the command

syst imp opta

conn

Initial values of states and parameters are assigned by

init xl: 1
init pil:2
nt pi2:2
The sampling period of the system OPTA is set by the command

par tinc:10
The command

par prin:5
tells that parameters, function values and gradients are printed. The optimization
is then executed by

plot y conl (t)

axesh010v-.1
simu O 10000 I

.5

l'

4t

t','
78

APPENDIX D

Macros for Generating the Figures

for

This appendix gives Simnon macros

macro FIG2
"Generates
syst vdpol

F9"

generating the figures in the report

split 1 1
axesh020v-66
plot x y

store x y
init x: I
simu 0 20 -mark/bl
mark a 2.5 O

t
markall
mark v "State variables x
mark "Time

and

marked

and 2

END

continuous system VDPOL

"The van der PoI equation

state x

der dx dy
dY=x

dx=a*x* (b-y*y)
a:1
b:1

-y

END

l'
macro FIG3
syst vdpol
init x: L
store x y
simu 0 2lb7

par b:2
s

imu/ b2

split 2 I
axesh020v-66
show x /b1
79

show x
axes v
show v
show Y

lb2

-33
lbt

lb2

mark a 2.5 0
mark tl.T. lme t

mark a 2.5 6.5

mark "Time t

markall
mark v "State variable
mark a L 7.5
mark v "State variable

y
x

END

macro FIG4
"Phase plane
syst vdpol
init x: 1

for the van der Pol equation

split 1 1
axesh-44v-33
plot y(x)

simu 0 20
mark a 2.5 O
mark 'oState variable x

markall
mark v "State variable

END

macro FIGS

"Simulation of population dynamics 0

syst

20

popdyn

split 3 1
axesh080v01
par r:0.2
plot x

simu 0 80
mark a 2.5 O
mark "Generation k

markall
mark v "Population
nark a I 5.5
mark v "Population
mark a110
mark v "Population

x
x

\,
, lt

axes

par r:2.7O

simu
axes

par r:2.83
s

imu

END

80

rr,

''

discrete system

POPDIN

"Simple model for population dynamics

state x
new nx
time k
tsamp ts
nx=x+r+x* (1-")
ts=k+1
x: 0.5

r:2

END

macro FIG6

syst POL
store f
simu -3.6 3.6
split 1 1

ashow f
mark a 17 0
mark ttx
mark a O.5 12
mark "f(x)
END

discrete system POL

time x
tsamp z
= (x+3) r (x+2) +x* (x-21* (x-3)
z=x+dx
dx: 0.05
END

macro FIGZ
"Generates

syst

Fig.

EXPZ

split 1 1
axesh-11v01.5
plot ImG (ReC)
simu 0 3.14
par alfa:135
simu 0 4.44
mark a 17 0
mark "Re G
mark a O.5
mark "Im G

1t

It

72

END

discrete system EXPZ

time r
tsamp

f
I

fi=alfa*pi/ 180
x=r* cos (fi )
81

(f i)
ReG=exp (x )*" os (v)
ImG=exp (x )*t in (v)

y=r* s in
5=*dr^

dr:0.01

pi:3.

t'4L5926

alfa:

90

END

macro FIGO
"Generates

Fig.

syst Proc Pireg-con

yr Y[Proc] tP'
"io."simu O 40/wuP
par ulow: -0.1
par uhigh:0.1
simu /nowuP

split 2

ashow Y/vruP
show Yr Y/nowuP
ashow uPr/wuP
show uPr/nowup

mark a 2"5 O
mark "Time t

markall
*u.f. "Control variable u
mark "a 7 7-5
rurf " "OutPut Y and Yref
END

continuous sYstem

;;;;;;;;;o'

PROC

*itt' inPut

saturation

input u
output Y
state x

ffii=i

u<-0.1 then -0'1 else

if

u<0'1 then u else 0'1

dx=upr
V=x

END

, t'

discrete

sYstem PIREG

input Yr
output u

iil-t"guf aior with anti-winduP

state i
new ni
time t

.,f
t
l'" '

tsamP ts

e=yr-y
9=l s+i
u=if vculow then ulow else
82

if v<uhigh then v else

uhigh

ni=i+k* e*h/ t i+u-v

ts=t+h
k:1

ti:

h: 0.5
ulow: -1

uhigh:

END

connecting system COll

"Connecting system for simulation of process
"with PI regulation by system PIREG

PROC

time t
YrIpireg]=1

y[Rireg ] =yIproc
uIproc ] =uIpi reg

]
]

END

macro FIG10

"l{acro for generating Fig.

let n.funcl=

10

syst FUNCI FUNCPLOT

get funcpar
split 1 1

axesh-44v-33
plot yp
simu -4 4
mark a 17 0
mark ttx
mark a I 12
mark "f (x)
END

connecting system

FUNCPLOT

time

x
u[FUNCl ]=x
yp=yIFUNCt]
END

IFUNC1]

UI1: -3.
UIZ: -L.
UI3: 1.

\
l,

't

UI4:4.
GI1:-1.
GI2: -2.8
GI3: 1.5

Gl(:2.

order: 1.
IFUNCPLoT]

83

APPENDIX

lntrac and Simnon Commands

This appendix lists all the eommands in Intrac and Simnon

INTRAC COMMANDS

1. Input and output

READ

WRITE
SWITCH
STOP

- Read variable from keyboard

- Write variables
- Utility command
- Stop execution and return to OS

2. Assignment

FREE
LET

Releases assigned global variables

Assigns global variables

3. Control of program flow

LABEL L - declaration of label
GOTO L
- transfer control
IF..GOTO - Transfer control
FOR..TO
NEXT V

Loop

4. Macro

Assigns default values

Macro definition
Declaration of formal 7
"n.r."rt,
'
End of macro definitiory'
Suspend execution of macro
Resume execution of macro

DEFAULT
MACRO
FORMAL
END
SUSPEND
RESUME

SIMNON COMMANDS

1. Input and output

EDIT
DISP
84

Edit system description

Display parameters

GET
LIST
PRINT
SAVE
STOP

- Get parameter and initial values

- List files
- Print files
- Save parameter values and initial values in a file
- Stop

2. Graphic output
AREA
ASHOW
AXES
HCOPY

MARK
SHOW
SPLIT
TEXT

Select window on screen

Plot stored variables with automatic scaling

Draw axes
Hardcopy of the screen
Write text on axes of graphs
Plot stored variables
Split screen into windows
Transfer text string to graph

3. Simulation
ALGOR
ERROR

INIT

PAR
PLOT
SIMU
STATE
STORE
SYST

Select integration algorithm

Choose emor bound for integration routine
Change initial values of state variables
Change parameters
Choose variables to be plotted
Simulate
Special command for the integration routine DAS
Choose variables to be stored
Activate systems

4. Auxiliary

HELP - Gives guidance

NEWS - Gives news about Simnon
TURN - Set switches

\
It

,rr^

85

lrdex

advanced features, 36
algebraic loop, 45
aigebraic loops, 44
ALGOR, 38, 49, 85
AREA, 20,29,50, g5
arithmetic operators, 18
ASHOW,50, g5
AXES, 72,20, 28, 51., 85
Backus-normal form, I
BNF, 8
changing parameters, 14
combination of systems, 30
command, 7
command syntax, 8, 49
cornputer control, 3i

coN,

83

conformal maps, 25
connecting system, 7, 33, 65
connecting system., 30
continuous system, 6, 30, 33, 64
CSMP, 47
CSSL, 47

DAS, 38, 45
DEFAULT, 84
default values, 7
DELAY, 38, 66
DER, 18
difference equations, 22
differential equations, 10
discrete system, 6, 26,30, 33, 64 * tt
,
DISP, 14,20,29, 51, g4
documentation, 42
EDIT, T2, T9,51, 84
edit mode, 11
edit mode: how to leave it, 12
END, 84
"4 ,f
ERROR, 37,52, 85
r
| ''
experimental data, 69
experimental data in simulations, 69
expression power, 38
expressions, 18
86

EXPZ, 81

field plots, 25
FIGIO, 83

FlGz,79
FIG3, 79
FIG4, BO
FIGs, BO
FIG6, 81
FIG7, 81
FiGg, 36, 82
file, 69
FOR..TO, 84

FORMAL,84

Fortran, 38
Fortran systems,

38

FREE, 84

FUNCI,38,39, 41,66,68
FUNCPLOT, 41, 83

functions, 19
functions in tabular form, 68
GET, 41, 52,85
global parameter, 39
global variables, 39
GOTO L, 84
graphs of functions, 24
HAMPC,38

hardcopy,

15

HCOPY, 15,20,28,52,95
HELP, 20, 53, 85
IDENTIFIERS, 63
IF..GOTO, 84
IFILE, 38, 66, 69
INIT, 12, 20,29,54, g5
integration algorithms, 37
interaction principles, 7
Intrac, 45
LABEL L, 84

large simulations,
LEAVE, 12
LET, 54, 84
LET,TERS, 63

41

level curves,25
LIST, 12, 19, 55, g5
local parameter, 39
local variables, 33, 39
LOGGER, 38, 66, 70
logical operators, 18

lower case letters, t6,

sToRE, 15,20,29,60,95
store parameters, 41
subsystem, 30
SUSPEND, 84
SWITCH, 42,60,84
syntax, 8, 49
syntax diagnam, 7
SYST, 72,20,29, 37,61, g5
system description, 10, 64
table nonlinearities, 38
TEXT, 61, g5
TIME, 18
time delay, 66
timing of simulations, 73

37

LP, 56
MACRO, 36,94
macros, 36
Macros for generating the figures, 79
MARK, 42,56,85
minimization, T4
NEWS, 56, 85
NEXT V, 84

TSAMP,23
TURN, 42,6I,85
upper case letters, 16
van der Pol equation, 10
variables, 11, 63
variables local, 33
variables transmission to other systems, 33
VDPOL, 79
vector notation, 19, 38
WRITE,84
[FUNCl], 83

NOISE, 38

'IOISEI, 66,72
numeric integration, 37

operators, 18
oPTA, 39,66,74
optimization, TS
PAR, 11, 14,20,28, 56, g5
parameter adjustment, 75
parameters, 11
phase plane plots, 16
Pl regulator,32
PLOT, 12,20,29, 56, 85
POL, 81
POPDYN, 80
population dynamics, 22
PRINT, 56, 85
FROC, g2
READ, 84
REG, 82

relational operators,

18

RESUME, 84

retrieve parameters,

41

RK, 38
RKFIX, 38
SAVE, 41,57,85
scale factors, 42

sHow, 20,29,57,85

Simnon commands, 84
Simnon language, 38
slMU, 7, 72, 20, 29, 59, g5
simulation, 12.
simulation command, 7

sorting, 43
source code,46
SPLIT, 20,2g,5g, g5
standard systems, use of,
STATE, 18,59,85
STIME, 66, 73
sToP, 60, 94, g5

|
, I'

,t

."4 f

39 | ',

87