~ubuntu-branches/ubuntu/trusty/yade/trusty

.def("makeCloud",&SpherePack::makeCloud,(python::arg("minCorner")=Vector3r(Vector3r::Zero()),python::arg("maxCorner")=Vector3r(Vector3r::Zero()),python::arg("rMean")=-1,python::arg("rRelFuzz")=0,python::arg("num")=-1,python::arg("periodic")=false,python::arg("porosity")=0.65,python::arg("psdSizes")=vector<Real>(),python::arg("psdCumm")=vector<Real>(),python::arg("distributeMass")=false,python::arg("seed")=0,python::arg("hSize")=Matrix3r(Matrix3r::Zero())),"Create random loose packing enclosed in a parallelepiped (also works in 2D if minCorner[k]=maxCorner[k] for one coordinate)."

"\nSphere radius distribution can be specified using one of the following ways:\n\n#. *rMean*, *rRelFuzz* and *num* gives uniform radius distribution in *rMean*×(1 ± *rRelFuzz*). Less than *num* spheres can be generated if it is too high.\n#. *rRelFuzz*, *num* and (optional) *porosity*, which estimates mean radius so that *porosity* is attained at the end. *rMean* must be less than 0 (default). *porosity* is only an initial guess for the generation algorithm, which will retry with higher porosity until the prescibed *num* is obtained.\n#. *psdSizes* and *psdCumm*, two arrays specifying points of the `particle size distribution <http://en.wikipedia.org/wiki/Particle_size_distribution>`__ function. As many spheres as possible are generated.\n#. *psdSizes*, *psdCumm*, *num*, and (optional) *porosity*, like above but if *num* is not obtained, *psdSizes* will be scaled down uniformly, until *num* is obtained (see :yref:`appliedPsdScaling<yade._packSpheres.SpherePack.appliedPsdScaling>`).\n\nBy default (with ``distributeMass==False``), the distribution is applied to particle radii. The usual sense of \"particle size distribution\" is the distribution of *mass fraction* (rather than particle count); this can be achieved with ``distributeMass=True``."

"\nSphere radius distribution can be specified using one of the following ways:\n\n#. *rMean*, *rRelFuzz* and *num* gives uniform radius distribution in *rMean×(1 ± rRelFuzz)*. Less than *num* spheres can be generated if it is too high.\n#. *rRelFuzz*, *num* and (optional) *porosity*, which estimates mean radius so that *porosity* is attained at the end. *rMean* must be less than 0 (default). *porosity* is only an initial guess for the generation algorithm, which will retry with higher porosity until the prescibed *num* is obtained.\n#. *psdSizes* and *psdCumm*, two arrays specifying points of the `particle size distribution <http://en.wikipedia.org/wiki/Particle_size_distribution>`__ function. As many spheres as possible are generated.\n#. *psdSizes*, *psdCumm*, *num*, and (optional) *porosity*, like above but if *num* is not obtained, *psdSizes* will be scaled down uniformly, until *num* is obtained (see :yref:`appliedPsdScaling<yade._packSpheres.SpherePack.appliedPsdScaling>`).\n\nBy default (with ``distributeMass==False``), the distribution is applied to particle radii. The usual sense of \"particle size distribution\" is the distribution of *mass fraction* (rather than particle count); this can be achieved with ``distributeMass=True``."

"\n\nIf *num* is defined, then sizes generation is deterministic, giving the best fit of target distribution. It enables spheres placement in descending size order, thus giving lower porosity than the random generation."

"\n\n:param Vector3 minCorner: lower corner of an axis-aligned box\n:param Vector3 maxCorner: upper corner of an axis-aligned box\n:param Matrix3 hSize: base vectors of a generalized box (arbitrary parallelepiped, typically :yref:`Cell::hSize`), superseeds minCorner and maxCorner if defined. For periodic boundaries only.\n:param float rMean: mean radius or spheres\n:param float rRelFuzz: dispersion of radius relative to rMean\n:param int num: number of spheres to be generated. If negavite (default), generate as many as possible with stochastic sizes, ending after a fixed number of tries to place the sphere in space, else generate exactly *num* spheres with deterministic size distribution.\n:param bool periodic: whether the packing to be generated should be periodic\n:param float porosity: initial guess for the iterative generation procedure (if *num*>1). The algorithm will be retrying until the number of generated spheres is *num*. The first iteration tries with the provided porosity, but next iterations increase it if necessary (hence an initialy high porosity can speed-up the algorithm). If *psdSizes* is not defined, *rRelFuzz* ($z$) and *num* ($N$) are used so that the porosity given ($\\rho$) is approximately achieved at the end of generation, $r_m=\\sqrt[3]{\\frac{V(1-\\rho)}{\\frac{4}{3}\\pi(1+z^2)N}}$. The default is $\\rho$=0.5. The optimal value depends on *rRelFuzz* or *psdSizes*.\n:param psdSizes: sieve sizes (particle diameters) when particle size distribution (PSD) is specified\n:param psdCumm: cummulative fractions of particle sizes given by *psdSizes*; must be the same length as *psdSizes* and should be non-decreasing\n:param bool distributeMass: if ``True``, given distribution will be used to distribute sphere's mass rather than radius of them.\n:param seed: number used to initialize the random number generator.\n:returns: number of created spheres, which can be lower than *num* depending on the method used.\n")

.def("psd",&SpherePack::psd,(python::arg("bins")=50,python::arg("mass")=true),"Return `particle size distribution <http://en.wikipedia.org/wiki/Particle_size_distribution>`__ of the packing.\n:param int bins: number of bins between minimum and maximum diameter\n:param mass: Compute relative mass rather than relative particle count for each bin. Corresponds to :yref:`distributeMass parameter for makeCloud<yade.pack.SpherePack.makeCloud>`.\n:returns: tuple of ``(cumm,edges)``, where ``cumm`` are cummulative fractions for respective diameters and ``edges`` are those diameter values. Dimension of both arrays is equal to ``bins+1``.")

Older »