ICODE-ADC/lammps
4
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity

git-svn-id: svn://svn.icms.temple.edu/lammps-ro/trunk@281 f3b2605a-c512-4ea7-a41b-209d697bcdaa

Browse Source
This commit is contained in:
sjplimp
2007-02-09 21:37:30 +00:00
parent 6e94c3652b
commit 47b87458b0
78 changed files with 3062 additions and 2395 deletions
Download Patch File Download Diff File Expand all files Collapse all files

217
doc/variable.txt
View File

@ -13,20 +13,23 @@ variable command :h3
variable name style args ... :pre
name = name of variable to define :ulb,l
style = {index} or {loop} or {equal} or {world} or {universe} :l
style = {index} or {loop} or {world} or {universe} or {uloop} or {equal} :l
{index} args = one or more strings
{loop} args = N = integer size of loop
{equal} args = one string containing functions, vectors, keywords, numbers
{world} args = one string for each partition of processors
{universe} args = one or more strings
{uloop} args = N = integer size of loop
{equal} args = one equation containing numbers, thermo keywords, math functions, group functions, atom vectors, compute references, other variables
numbers = 0.0, -5.4, 2.8e-4, etc
thermo keywords = vol, ke, press, etc from "thermo_style"_thermo_style.html
math functions = add(x,y), sub(x,y), mult(x,y), div(x,y),
neg(x), pow(x,y), exp(x), ln(x), sqrt(x)
group functions = mass(group), charge(group), xcm(group,dim),
vcm(group,dim), bound(group,xmin), gyration(group)
vectors = x\[5\], y\[12\], z\[17\], vx\[88\], vy\[19\], vz\[2\],
fx\[1\], fy\[2005\], fz\[1\]
keywords = same keywords (mostly) as in "thermo_style custom"_thermo_style.html command
{world} args = one string for each partition of processors
{universe} args = one or more strings
{uloop} args = N = integer size of loop :pre
atom vectors = x\[N\], y\[N\], z\[N\], vx\[N\], vy\[N\], vz\[N\],
fx\[N\], fy\[N\], fz\[N\]
compute references = c_ID\[0\], c_ID\[N\]
other variables = v_abc, v_x, etc :pre
:ule
[Examples:]
@ -34,8 +37,9 @@ style = {index} or {loop} or {equal} or {world} or {universe} :l
variable x index run1 run2 run3 run4 run5 run6 run7 run8
variable LoopVar loop 20
variable beta equal div(temp,3.0)
variable b1 equal add(x\[234\],mult(0.5,lx))
variable b equal xcm(mol1,x)
variable b1 equal add(x\[234\],mult(0.5,col))
variable b equal div(xcm(mol1,x),2.0)
variable b equal c_myTemp\[0\]
variable temp world 300.0 310.0 320.0 330.0
variable x universe 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15
variable x uloop 15 :pre
@ -43,15 +47,24 @@ variable x uloop 15 :pre
[Description:]
This command assigns one or more values to a variable name so that the
variable can be used in subsequent input script commands. The "name"
of the variable is an arbitrary string. Each "value" is a string
which could be text or numbers, as in the examples above. As
explained in "this section"_Section_commands.html#3_2, occurrences of
the variable name in an input script line are replaced by the
variable can be used in subsequent input script commands or its value
output during a simulation. The "name" of the variable is an
arbitrary string. Each "value" is a string which could be text or
numbers, as in the examples above.
As explained in "this section"_Section_commands.html#3_2, occurrences
of the variable name in an input script line are replaced by the
variable's value. The variable name can be referenced in the input
script as $x if the name "x" is a single character, or as $\{LoopVar\}
if the name "LoopVar" is one or more characters.
Variable values can also be accessed for output once or periodically
during a simulation by the "print"_print.html command, "fix
print"_fix_print.html command, "run every"_run.html command, and the
"thermo_style"_thermo_style.html command.
:line
As described below, for variable styles {index}, {loop}, {universe},
and {uloop}, the value assigned to a variable can be incremented via
the "next"_next.html command. When there are no more values to
@ -73,13 +86,15 @@ equal-style variable is encountered. Also, if a variable is iterated
on to the end of its list via the "next"_next.html command, it is
available to be re-defined in a subsequent variable command.
:line
For the {index} style, one or more strings are specified. Initially,
the 1st string is assigned to the variable. Each time a
"next"_next.html command is used with the variable name, the next
string is assigned. All processors assign the same string to the
variable. {Index}-style variables can also be set (with a single
value) by using the command-line switch -var; see "this
section"_Section_start.html#2_4 for details.
section"_Section_start.html#2_6 for details.
The {loop} style is identical to the {index} style except that the
strings are the integers from 1 to N. This allows you to generate a
@ -89,66 +104,9 @@ Each time a "next"_next.html command is used with the variable name,
the next string ("2", "3", etc) is assigned. All processors assign
the same string to the variable.
For the {equal} style, a single string is specified which represents
an equation that will be evaluated afresh each time the variable is
used. Thus the variable can take on different values at different
stages of the input script. For example, if the variable is used in a
"fix print"_fix_print.html command, it could print different values
each timestep it was invoked. The next command cannot be used with
{equal}-style variables, since there is only one value. Note that, as
with any other input script command, it is feasible to use another
variable in the {equal} variable's string, e.g. variable y equal
mult($x,2). However, $x will be replaced immediately by it's current
value when the command is first parsed, not each time that $y is
substituted for.
The syntax of the equation assigned to {equal} variables is simple.
It can contain "functions", "vectors", "keywords", or "numbers" in any
combination.
Function = a keyword followed by parenthesis with one or two arguments
Supported arithmetic functions = add(x,y), sub(x,y), mult(x,y), div(x,y), \
neg(x), pow(x,y), exp(x), ln(x), sqrt(x)
Supported group functions = mass(ID), charge(ID), xcm(ID,dim), vcm(ID,dim), \
bound(ID,dir), gyration(ID)
Example function usage = div(1.0e20,3.0), neg(x\[34\]), pow(lx,3.0), \
xcm(mol,x), bound(lower,zmin)
Vector = a keyword followed by square brackets containing an atom ID
Supported vectors = x, y, z, vx, vy, vz, fx, fy, fz
Example vector usage = x\[123\], fz\[1000\]
Keyword = any keyword supported by the \
"thermo_style custom"_thermo_style.html \
command except cpu, pressure tensor components (pxx, pyy, etc), \
time-averaged quantities (tave, pave, etc)
Example keyword usage = atoms, pow(vol,0.333), mult(elong,0.5)
Number = 0.2, 1.0e20, -15.4, etc :ul
For the group functions, ID is a group-ID, dim is 'x' or 'y' or 'z',
and dir is one of 6 strings: "xmin", "xmax", "ymin", "ymax", "zmin",
or "zmax". The group functions mass() and charge() are the total mass
and charge of the group of atoms. Xcm() and vcm() return components
of the position and velocity of the center of mass of the group.
Bound() returns the min/max of a particular coordinate for all atoms
in the group. Gyration() computes the radius-of-gyration of the group
of atoms. See the "fix gyration"_fix_gyration.html command for the
formula.
Keywords have restrictions on when they can be assigned to variables.
For example, keywords that compute thermodynamic quantites can only be
invoked after the first simulation has begun. A warning is issued if
thermodyanmic keywords are invoked on timesteps when thermodynamic
information is not being printed to the screen, since the values
assigned to the variable may be out-of-date.
The variable {equal} equation can also be nested in that function
arguments can be functions, vectors, keywords, or numbers. For
example, this is a valid equation:
variable x equal div(add(pe,ke),pow(vol,div(1,3))) :pre
For the {world} style, one or more strings are specified. There must
be one string for each processor partition or "world". See "this
section"_Section_start.html#2_4 of the manual for information on
section"_Section_start.html#2_6 of the manual for information on
running LAMMPS with multiple partitions via the "-partition"
command-line switch. This variable command assigns one string to each
world. All processors in the world are assigned the same string. The
@ -161,7 +119,7 @@ different partitions.
For the {universe} style, one or more strings are specified. There
must be at least as many strings as there are processor partitions or
"worlds". See "this page"_Section_start.html#2_4 for information on
"worlds". See "this page"_Section_start.html#2_6 for information on
running LAMMPS with multiple partitions via the "-partition"
command-line switch. This variable command initially assigns one
string to each world. When a "next"_next.html command is encountered
@ -179,11 +137,118 @@ strings are the integers from 1 to N. This allows you to generate a
long list of runs (e.g. 1000) without having to list N values in your
input script.
:line
For the {equal} style, a single string is specified which represents
an equation that will be evaluated afresh each time the variable is
used. Thus the variable can take on different values at different
stages of the input script. For example, if the variable is used in a
"fix print"_fix_print.html command, different values could be printed
each timestep it was invoked. The next command cannot be used with
{equal}-style variables, since there is only one value.
The equation for an {equal}-style variable can contain a variety of
quantities. The syntax for each kind of quantity is simple, but
multiple quantities can be nested and combined in various ways to
build up formulas of arbitrary complexity. For example, this is a
valid (though strange) variable equation:
variable x equal div(add(pe,c_MyTemp\[0\]),pow(vol,div(1,3))) :pre
Specifically, an equation can contain numbers, thermo keywords, math
functions, group functions, atom vectors, compute references, and
other variables:
Number: 0.2, 1.0e20, -15.4, etc
Thermo keywords: vol, pe, ebond, etc
Math functions: add(x,y), sub(x,y), mult(x,y), div(x,y), \
neg(x), pow(x,y), exp(x), ln(x), sqrt(x)
Group functions: mass(ID), charge(ID), xcm(ID,dim), vcm(ID,dim), \
bound(ID,dir), gyration(ID)
Atom vectors: x\[N\], y\[N\], z\[N\], vx\[N\], vy\[N\], vz\[N\], \
fx\[N\], fy\[N\], fz\[N\]
Compute references: c_ID\[0\], c_ID\[N\]
Other variables: v_abc, v_x, etc :tb(s=:)
The thermo keywords allowed in the equation are those defined by the
"thermo_style custom" command. Note that many thermodyanmic
quantities are only computable after the first simulation has begun.
Likewise, many thermodynamic quantities (such as energies) are only
computed on timesteps when thermodyanmic output is being performed.
If the variable equation these quantities at other times, out-of-date
or invalid values may be used.
Math functions take one or two arguments, each of which may be an
equation containing any of the quantities defined above. This allows
equations to be nested, as in the examples above.
Group functions take one or two arguments. The first argument is the
group-ID. The {dim} argument is {x} or {y} or {z}. The {dir}
argument is {xmin}, {xmax}, {ymin}, {ymax}, {zmin}, or {zmax}. The
group functions mass() and charge() are the total mass and charge of
the group of atoms. Xcm() and vcm() return components of the position
and velocity of the center of mass of the group. Bound() returns the
min/max of a particular coordinate for all atoms in the group.
Gyration() computes the radius-of-gyration of the group of atoms. See
the "fix gyration"_fix_gyration.html command for the formula.
The atom vectors take a single integer argument from 1-N, which
is the desired atom-ID, e.g. x\[243\].
Compute references access allow access to scalar or vector quantities
calculated by a compute. The ID in the reference should be replaced
by the actual ID of the compute that has been defined elsewhere in the
input script. See the "compute"_compute.html command for details.
Note that per-atom quantities calcalated by a compute cannot be
accessed this way, but only global scalar or vector quantities.
If {c_ID\[0\]} is used as a keyword, then the scalar quantity
calculated by the compute is printed. If {c_ID\[N\]} is used, then N
in the range from 1-M will print a specific component of the vector
calculated by the compute.
The current values of other variables can be accessed by prepending a
"v_" to the variable name. This will cause the other variable to be
evaulated. Note that if you do something circular like this:
variable a equal v_b
variable b equal v_a
print $a :pre
then LAMMPS will run for a while when the print statement is invoked.
Note that there is a subtle difference between using a variable
in a {equal}-style equation in the form $x versus v_x.
In the former case, as with any other input script command, the
variable's value is substituted for immediately when the line is read
from the input script. Thus if the current simulation box volume was
1000.0, then these lines:
variable x equal vol
variable y equal mult($x,2) :pre
would associate the equation string "mult(1000.0,2)" with variable y.
By contrast, these lines:
variable x equal vol
variable y equal mult(v_x,2) :pre
would associate the equation string "mult(v_x,2)" with variable y.
Thus if the variable y were evaluated periodically during a run where
the box volume changed, the resulting value would always be 500.0 for
the first case, but would change dynamically for the second case.
:line
[Restrictions:]
The use of atom vectors in {equal} style variables requires the atom
style to use a global mapping in order to look up the vector indices.
Only atom styles with molecular information create global maps.
Only atom styles with molecular information create global maps unless
the "atom_modify map"_atom_modify.html command is used.
[Related commands:]