Syntax:
fix ID ave/grid group-ID Nevery Nrepeat Nfreq value1 value2 ... keyword args ...
c_ID = per-grid vector calculated by a compute with ID c_ID[N] = Nth column of per-grid array calculated by a compute with ID, N can include wildcard (see below) f_ID = per-grid vector calculated by a fix with ID f_ID[N] = Nth column of per-grid array calculated by a fix with ID, N can include wildcard (see below) v_name = per-grid vector calculated by a grid-style variable with name g_name = custom per-grid vector with name g_name[N] = Nth column of per-grid custom array with name, N can include wildcard (see below)
keyword = ave
ave args = one or running
one = output a new average value every Nfreq steps
running = accumulate average continuously
Examples:
fix 1 ave/grid all 10 20 1000 c_mine fix 1 ave/grid all 1 100 100 c_2[1] ave running fix 1 ave/grid all 1 100 100 c_2[*] ave running fix 1 ave/grid section1 5 20 100 v_myEng
These commands will dump averages for each species and each grid cell to a file every 1000 steps:
compute 1 grid species n u v w usq vsq wsq fix 1 ave/grid 10 100 1000 c_1[*] dump 1 grid all 1000 tmp.grid id f_1[*]
Description:
Use one or more per-grid vectors as inputs every few timesteps, and average by grid cell over longer timescales, applying appropriate normalization factors. The resulting per grid cell averages can be used by other output commands such as the dump grid command. Only grid cells in the grid group specified by group-ID are included in the averaging. See the group grid command for info on how grid cells can be assigned to grid groups.
Each input value can be the result of a compute or fix or grid-style variable. The compute or fix must produce a per-grid vector or array, not a global or per-particle or per-surf quantity. If you wish to time-average global quantities from a compute, fix, or variable, then see the fix ave/time command. To time-average per-surf quantities, see the fix ave/surf command.
Each per-grid value of each input vector is averaged independently.
Computes that produce per-grid vectors or arrays are those which have the word grid in their style name. See the doc pages for individual fixes to determine which ones produce per-grid vectors or arrays.
Note that for values from a compute or fix or custom attribute, the bracketed index can be specified using a wildcard asterisk with the index to effectively specify multiple values. This takes the form "*" or "*n" or "n*" or "m*n". If N = the size of the vector (for mode = scalar) or the number of columns in the array (for mode = vector), then an asterisk with no numeric values means all indices from 1 to N. A leading asterisk means all indices from 1 to n (inclusive). A trailing asterisk means all indices from n to N (inclusive). A middle asterisk means all indices from m to n (inclusive).
Using a wildcard is the same as if the individual columns of the array had been listed one by one. E.g. these 2 fix ave/grid commands are equivalent, since the compute grid command creates a per-grid array with 3 columns:
compute myGrid all all u v w fix 1 ave/grid all 10 20 1000 c_myGrid[*] fix 1 ave/grid all 10 20 1000 c_myGrid[1] c_myGrid[2] c_myGrid[3]
The Nevery, Nrepeat, and Nfreq arguments specify on what timesteps the input values will be used in order to contribute to the average. The final averaged quantities are generated on timesteps that are a multiple of Nfreq. The average is over Nrepeat quantities, computed in the preceding portion of the simulation every Nevery timesteps. Nfreq must be a multiple of Nevery and Nevery must be non-zero even if Nrepeat is 1. Also, the timesteps contributing to the average value cannot overlap, i.e. Nfreq > (Nrepeat-1)*Nevery is required.
For example, if Nevery=2, Nrepeat=6, and Nfreq=100, then values on timesteps 90,92,94,96,98,100 will be used to compute the final average on timestep 100. Similarly for timesteps 190,192,194,196,198,200 on timestep 200, etc.
If a value begins with "c_", a compute ID must follow which has been previously defined in the input script. If no bracketed term is appended, the compute must calculate a per-grid vector. If c_ID[N] is used, the compute must calculate a per-grud array with M columns and N must be in the range from 1-M, which will use the Nth column of the M-column per-grid array. See the discussion above for how N can be specified with a wildcard asterisk to effectively specify multiple values.
Users can also write code for their own compute styles and add them to SPARTA.
If a value begins with "f_", a fix ID must follow which has been previously defined in the input script. If no bracketed term is appended, the fix must calculates a per-grid vector. If f_ID[N] is used, the fix must calculate a per-grid array with M columns and N must be in the range from 1-M, which will use the Nth column of the M-column per-grid array. See the discussion above for how N can be specified with a wildcard asterisk to effectively specify multiple values.
Note that some fixes only produce their values on certain timesteps, which must be compatible with Nevery, else an error will result. Users can also write code for their own fix styles and add them to SPARTA.
If a value begins with "v_", a variable name must follow which has been previously defined in the input script. Only grid-style variables can be referenced. See the variable command for details. Note that grid-style variables define a formula which can reference stats_style keywords, or they can invoke other computes, fixes, or variables when they are evaluated, so this is a very general means of specifying quantities to time average.
If a value begins with "g_", the name of a custom per-grid vector or array must follow. Custom attributes can store either a single or multiple values per grid cell. See Section 6.17 for more discussion of custom attributes and command that define them. For example, the read_grid and surf_react implicit commands can define per-grid attributes. (The surf/react implicit command has not yet been released in public SPARTA).
If g_name is used as a value, the custom attribute must be a vector. If g_name[N] is used, the custom attribute must be an array, and N must be in the range from 1-M for an M-column array. See the discussion above for how N can be specified with a wildcard asterisk to effectively specify multiple values.
For averaging of a value that comes from a compute or fix, normalization is performed as follows. Note that no normalization is performed on a value produced by a grid-style variable.
If the compute or fix is summing over particles in a grid cell to calculate a per-grid quantity (e.g. energy or temperature), this takes the form of a numerator divided by a denominator. For example, see the formulas discussed on the compute grid doc page, where the denominator is 1 (for keyword n), or the number of particles (ke, mass, temp), or the sum of particle masses (u, usq, etc). When this command averages over a series of timesteps, the numerator and denominator are summed separately. This means the numerator/denominator division only takes place when this fix produces output, every Nfreq timesteps.
For example, say the Nfreq output is over 2 timesteps, and the value produced by compute grid mass is being averaged. Say a grid cell has 10 particles on the 1st timestep with a numerator value of 10.0, and 100 particles on the 2nd timestep with a numerator value of 50.0. The output of this fix will be (10+50) / (10+100) = 0.54, not ((10/10) + (50/100)) / 2 = 0.75.
Additional optional keywords also affect the operation of this fix.
The ave keyword determines what happens to the accumulation of statistics every Nfreq timesteps.
If the ave setting is one, then the values produced on timesteps that are multiples of Nfreq are independent of each other. Normalization as described above is performed, and all tallies are zeroed before accumulating over the next Nfreq steps.
If the ave setting is running, then tallies are never zeroed. Thus the output at any Nfreq timestep is normalized over all previously accumulated samples since the fix was defined. The tallies can only be zeroed by deleting the fix via the unfix command, or by re-defining the fix, or by re-specifying it.
Restart, output info:
No information about this fix is written to binary restart files.
This fix produces a per-grid vector or array which can be accessed by various output commands. A vector is produced if only a single quantity is averaged by this fix. If two or more quantities are averaged, then an array of values is produced, where the number of columns is the number of quantities averaged. The per-grid values can only be accessed on timesteps that are multiples of Nfreq since that is when averaging is performed.
This fix performs averaging for all child grid cells in the simulation, which includes unsplit, split, and sub cells. Section 4.8 of the manual gives details of how SPARTA defines child, unsplit, split, and sub cells.
Grid cells not in the specified group-ID will output zeroes for all their values.
Styles with a kk suffix are functionally the same as the corresponding style without the suffix. They have been optimized to run faster, depending on your available hardware, as discussed in the Accelerating SPARTA section of the manual. The accelerated styles take the same arguments and should produce the same results, except for different random number, round-off and precision issues.
These accelerated styles are part of the KOKKOS package. They are only enabled if SPARTA was built with that package. See the Making SPARTA section for more info.
You can specify the accelerated styles explicitly in your input script by including their suffix, or you can use the -suffix command-line switch when you invoke SPARTA, or you can use the suffix command in your input script.
See the Accelerating SPARTA section of the manual for more instructions on how to use the accelerated styles effectively.
Restrictions:
If one of the specified values is a compute which tallies information on collisions between particles and implicit surface element within each grid cell, then all the values must be for compute(s) which do this. I.e. you cannot mix computes which operate on implicit surfaces with other kinds of per-grid values in the same fix ave/grid command.
Examples of computes which tally particle/implicit surface element collision info within each grid cell are compute isurf/grid and compute react/isurf/grid.
If performing on-the-fly grid adaptation every N timesteps, using the fix adapt command, this fix cannot time-average across time windows > N steps, since the grid may change. This means Nfreq cannot be > N, and keyword ave = running is not allowed.
Related commands:
Default:
The option defaults are ave = one.