source: inundation/pyvolution/advection.py @ 3354

import sys
from os import sep
sys.path.append('..'+sep+'pyvolution')

"""Class Domain -
2D triangular domains for finite-volume computations of
the advection equation.

This module contains a specialisation of class Domain from module domain.py
consisting of methods specific to the advection equantion

The equation is

  u_t + (v_1 u)_x + (v_2 u)_y = 0

There is only one conserved quantity, the stage u

The advection equation is a very simple specialisation of the generic
domain and may serve as an instructive example or a test of other
components such as visualisation.

Ole Nielsen, Stephen Roberts, Duncan Gray, Christopher Zoppou
Geoscience Australia, 2004
"""


import logging, logging.config
logger = logging.getLogger('advection')
logger.setLevel(logging.WARNING)

try:
    logging.config.fileConfig('log.ini')
except:
    pass


from domain import *
Generic_domain = Domain #Rename

class Domain(Generic_domain):

    def __init__(self,
                 coordinates,
                 vertices,
                 boundary = None,
                 tagged_elements = None,
                 geo_reference = None,
                 use_inscribed_circle=False,
                 velocity = None,
                 full_send_dict=None,
                 ghost_recv_dict=None,
                 processor=0,
                 numproc=1
                 ):

        conserved_quantities = ['stage']
        other_quantities = []
        Generic_domain.__init__(self,
                                source=coordinates,
                                triangles=vertices,
                                boundary=boundary,
                                conserved_quantities=conserved_quantities,
                                other_quantities=other_quantities,
                                tagged_elements=tagged_elements,
                                geo_reference=geo_reference,
                                use_inscribed_circle=use_inscribed_circle,
                                full_send_dict=full_send_dict,
                                ghost_recv_dict=ghost_recv_dict,
                                processor=processor,
                                numproc=numproc)


        if velocity is None:
            self.velocity = [1,0]
        else:
            self.velocity = velocity

        #Only first is implemented for advection
        self.default_order = self.order = 1

        #Realtime visualisation
        self.visualiser = None
        self.visualise  = False
        self.visualise_color_stage = False
        self.visualise_stage_range = 1.0
        self.visualise_timer = True
        self.visualise_range_z = None


        self.smooth = True

    def initialise_visualiser(self,scale_z=1.0,rect=None):
        #Realtime visualisation
        if self.visualiser is None:
            from realtime_visualisation_new import Visualiser
 #           from vtk_realtime_visualiser import Visualiser
            self.visualiser = Visualiser(self,scale_z,rect)
        self.visualise = True


    def check_integrity(self):
        Generic_domain.check_integrity(self)

        msg = 'Conserved quantity must be "stage"'
        assert self.conserved_quantities[0] == 'stage', msg


    def flux_function(self, normal, ql, qr, zl=None, zr=None):
        """Compute outward flux as inner product between velocity
        vector v=(v_1, v_2) and normal vector n.

        if <n,v> > 0 flux direction is outward bound and its magnitude is
        determined by the quantity inside volume: ql.
        Otherwise it is inbound and magnitude is determined by the
        quantity outside the volume: qr.
        """

        v1 = self.velocity[0]
        v2 = self.velocity[1]


        normal_velocity = v1*normal[0] + v2*normal[1]

        if normal_velocity < 0:
            flux = qr * normal_velocity
        else:
            flux = ql * normal_velocity

        max_speed = abs(normal_velocity)
        return flux, max_speed

    def compute_fluxes(self):

        try:
            import weave
            self.weave_available = True
        except:
            self.weave_available = False

        if self.weave_available:
            self.compute_fluxes_weave()
        else:
            self.compute_fluxes_python()



    def compute_fluxes_python(self):
        """Compute all fluxes and the timestep suitable for all volumes
        in domain.

        Compute total flux for each conserved quantity using "flux_function"

        Fluxes across each edge are scaled by edgelengths and summed up
        Resulting flux is then scaled by area and stored in
        domain.explicit_update

        The maximal allowable speed computed by the flux_function
        for each volume
        is converted to a timestep that must not be exceeded. The minimum of
        those is computed as the next overall timestep.

        Post conditions:
        domain.explicit_update is reset to computed flux values
        domain.timestep is set to the largest step satisfying all volumes.
        """

        import sys
        from Numeric import zeros, Float
        from config import max_timestep

        N = self.number_of_elements

        neighbours = self.neighbours
        neighbour_edges = self.neighbour_edges
        normals = self.normals

        areas = self.areas
        radii = self.radii
        edgelengths = self.edgelengths

        timestep = max_timestep #FIXME: Get rid of this

        #Shortcuts
        Stage = self.quantities['stage']

        #Arrays
        stage = Stage.edge_values

        stage_bdry = Stage.boundary_values

        flux = zeros(1, Float) #Work array for summing up fluxes

        #Loop
        for k in range(N):
            optimal_timestep = float(sys.maxint)

            flux[:] = 0.  #Reset work array
            for i in range(3):
                #Quantities inside volume facing neighbour i
                ql = stage[k, i]

                #Quantities at neighbour on nearest face
                n = neighbours[k,i]
                if n < 0:
                    m = -n-1 #Convert neg flag to index
                    qr = stage_bdry[m]
                else:
                    m = neighbour_edges[k,i]
                    qr = stage[n, m]


                #Outward pointing normal vector
                normal = normals[k, 2*i:2*i+2]

                #Flux computation using provided function
                edgeflux, max_speed = self.flux_function(normal, ql, qr)
                flux -= edgeflux * edgelengths[k,i]

                #Update optimal_timestep
                if  self.tri_full_flag[k] == 1 :
                    try:
                        optimal_timestep = min(optimal_timestep, radii[k]/max_speed)
                    except ZeroDivisionError:
                        pass

            #Normalise by area and store for when all conserved
            #quantities get updated
            flux /= areas[k]
            Stage.explicit_update[k] = flux[0]

            timestep = min(timestep, optimal_timestep)

        self.timestep = timestep

    def compute_fluxes_weave(self):
        """Compute all fluxes and the timestep suitable for all volumes
        in domain.

        Compute total flux for each conserved quantity using "flux_function"

        Fluxes across each edge are scaled by edgelengths and summed up
        Resulting flux is then scaled by area and stored in
        domain.explicit_update

        The maximal allowable speed computed by the flux_function
        for each volume
        is converted to a timestep that must not be exceeded. The minimum of
        those is computed as the next overall timestep.

        Post conditions:
        domain.explicit_update is reset to computed flux values
        domain.timestep is set to the largest step satisfying all volumes.
        """

        import sys
        from Numeric import zeros, Float
        from config import max_timestep

        import weave
        from weave import converters

        N = self.number_of_elements


        timestep    = zeros( 1, Float);
        timestep[0] = float(max_timestep) #FIXME: Get rid of this

        #Shortcuts
        Stage = self.quantities['stage']

        #Arrays
        neighbours      = self.neighbours
        neighbour_edges = self.neighbour_edges
        normals         = self.normals
        areas           = self.areas
        radii           = self.radii
        edgelengths     = self.edgelengths
        tri_full_flag   = self.tri_full_flag

        stage_edge      = Stage.edge_values
        stage_bdry      = Stage.boundary_values
        stage_update    = Stage.explicit_update

        huge_timestep = float(sys.maxint)

        v1 = self.velocity[0]
        v2 = self.velocity[1]

        code = """
        //Loop

        double qr,ql;
        int m,n;
        double normal[2];
        double normal_velocity;
        double flux, edgeflux;
        double max_speed;
        double optimal_timestep;
        for (int k=0; k<N; k++){

            optimal_timestep = huge_timestep;
            flux = 0.0;  //Reset work array
            for (int i=0; i<3; i++){
                //Quantities inside volume facing neighbour i
                ql = stage_edge(k, i);

                //Quantities at neighbour on nearest face
                n = neighbours(k,i);
                if (n < 0) {
                    m = -n-1; //Convert neg flag to index
                    qr = stage_bdry(m);
                } else {
                    m = neighbour_edges(k,i);
                    qr = stage_edge(n, m);
                }


                //Outward pointing normal vector
                for (int j=0; j<2; j++){
                    normal[j] = normals(k, 2*i+j);
                }


                //Flux computation using provided function
                normal_velocity = v1*normal[0] + v2*normal[1];

                if (normal_velocity < 0) {
                    edgeflux = qr * normal_velocity;
                } else {
                    edgeflux = ql * normal_velocity;
                }

                max_speed = fabs(normal_velocity);
                flux = flux - edgeflux * edgelengths(k,i);

                //Update optimal_timestep
                if (tri_full_flag(k) == 1) {
                    if (max_speed != 0.0) {
                        optimal_timestep = (optimal_timestep>radii(k)/max_speed) ? radii(k)/max_speed : optimal_timestep;
                    }
                }

            }

            //Normalise by area and store for when all conserved
            //quantities get updated
            stage_update(k) = flux/areas(k);

            timestep(0) = (timestep(0)>optimal_timestep) ? optimal_timestep : timestep(0);

        }
        """

        logger.debug('Trying to weave advection.compute_fluxes')
        weave.inline(code, ['stage_edge','stage_bdry','stage_update',
                             'neighbours','neighbour_edges','normals',
                             'areas','radii','edgelengths','tri_full_flag',
                             'huge_timestep',
                             'timestep','v1','v2','N'],
                             type_converters = converters.blitz, compiler='gcc');

        self.timestep = timestep[0]



    def evolve(self,
               yieldstep = None,
               finaltime = None,
               duration = None,
               skip_initial_step = False):

        """Specialisation of basic evolve method from parent class
        """

        #Initialise real time viz if requested
        if self.visualise is True and self.time == 0.0:
            #pass
            #import realtime_visualisation_new as visualise
            #visualise.create_surface(self)
            self.initialise_visualiser()
            self.visualiser.setup_all()
            self.visualiser.update_timer()


        #Call basic machinery from parent class
        for t in Generic_domain.evolve(self,
                                       yieldstep=yieldstep,
                                       finaltime=finaltime,
                                       duration=duration,
                                       skip_initial_step=skip_initial_step):




            #Real time viz
            if self.visualise is True:
                #pass
                self.visualiser.update_all()
                self.visualiser.update_timer()


            #Pass control on to outer loop for more specific actions
            yield(t)