3.7.8.2. Sweep Definitions

Each sweep has a definition that is similar to a report but with additional options to divide the run matrix into subsets. For example, if the run matrix has three independent variables (which pyCart calls trajectory keys) of "Mach", "alpha", and "beta", then a common sweep would be to plot results as a function of Mach number for constant alpha and beta. To do that, one would put "EqCons": ["alpha", "beta"] in the sweep definition.

The full list of available options is below.

Sweeps: {} | {[S]} | dict (dict)

Dictionary of sweep definitions (combined plots of subsets of cases)

S: dict

Dictionary of sweep definitions for sweep named "S"

Figures: {[]} | list (str)

List of figures to include for each sweep subset

EqCons: {[]} | list (str)

List of trajectory keys to hold constant for each subset

TolCons: {{}} | dict (float)

Dictionary of trajectory keys to hold within a certain tolerance from the value of that key for the first case in the subset

IndexTol: {None} | int

If used, only allow the index of the first and last cases in a subset to differ by this value

XAxis: {None} | str

Name of trajectory key used to sort subset; if None, sort by data book index

RunMatrixOnly: true | {false}

By default, the data book is the source for sweep plots; this option can restrict the plots to points in the current run matrix

GlobalCons: {[]} | list (str)

List of global constraints to only divide part of the run matrix into subsets

Indices: {None} | list (int)

If used, list of indices to divide into subsets

MinCases: {1} | int

Minimum number of cases for a sweep to be reported

CarpetEqCons: [] | list (str)

Some sweep subfigures allow a sweep to be subdivided into subsweeps; this could be used to create plots of CN versus Mach with several lines each having constant alpha

CarpetTolCons: {} | dict (float)

Tolerance constraints for subdividing sweeps

The subsets are defined so that each case meeting the GlobalCons is placed into exactly one subset. For each subset, pyCart begins with the first available case and applies the constraints using that point as a reference.

Constraints can be defined in more complex ways than the example given prior to the list of options. For relatively simple run matrices, grouping cases by constant values of one or more trajectory keys (i.e. using EqCons) may be adequate, but other run matrices may require more advanced settings.

For example, wind tunnel data often is collected at conditions that are not exactly constant, i.e. the angle of attack may fluctuate slightly. Instead of using EqCons, a better option in this case would be to include "TolCons": {"alpha": 0.02}. Then all cases in a subset would have an angle of attack within 0.02 of the angle of attack of the first point of the subset.

Another advanced capability is to use EqCons such as ["k%10"] or ["k/10%10"]. This could be used to require each case to have the same ones digit or the same tens digit of some trajectory variable called k.