def test_cut_bwb(self):
        """recover element ids"""
        log = SimpleLogger(level='warning', encoding='utf-8', log_func=None)
        is_bwb = True
        if is_bwb:
            bdf_filename = os.path.join(MODEL_PATH, 'bwb',
                                        'bwb_saero.bdf')  # ymax~=1262.0
        else:  #  pragma: no cover
            bdf_filename = r'C:\NASA\asm\all_modes_mach_0.85\flutter.bdf'  # ymax=1160.601
        normal_plane = np.array([0., 1., 0.])

        model = read_bdf(bdf_filename, log=log)
        model2 = read_bdf(bdf_filename, log=log)

        # initialize theta
        thetas = {}
        for eid in model.elements:
            #  theta, Ex, Ey, Gxy
            thetas[eid] = (0., 0., 0., 0.)

        #p1 = np.array([466.78845, 735.9053, 0.0])
        #p2 = np.array([624.91345, 639.68896, -0.99763656])
        #dx = p2 - p1
        ytol = 2.
        nodal_result = None
        plane_bdf_filenames = []
        y = []
        A = []
        I = []
        EI = []
        avg_centroid = []

        for i in range(2000):
            if is_bwb:
                dy = 100. * i + 1.  #  bwb
                coord = CORD2R(1,
                               rid=0,
                               origin=[0., dy, 0.],
                               zaxis=[0., dy, 1],
                               xzplane=[1., dy, 0.])
            else:  #  pragma: no cover
                dy = 4. * i + 1.  #  CRM
                coord = CORD2R(1,
                               rid=0,
                               origin=[0., dy, 0.],
                               zaxis=[0., dy, 1],
                               xzplane=[1., dy, 0.])
                #origin = np.array([0., dy, 0.])
                #xzplane = origin + dx
                #xzplane = np.array([1., dy, 0.])
                #coord = CORD2R.add_axes(cid, rid=0, origin=p1, xaxis=p2-p1, yaxis=None, zaxis=None,
                #xyplane=None, yzplane=None, xzplane=None, comment='')
                print(coord)
            model.coords[1] = coord
            plane_bdf_filename = 'plane_face_%i.bdf' % i
            cut_face_filename = 'cut_face_%i.csv' % i
            if os.path.exists(cut_face_filename):
                os.remove(cut_face_filename)
            try:
                out = cut_face_model_by_coord(
                    model2,
                    coord,
                    ytol,
                    nodal_result,
                    plane_atol=1e-5,
                    skip_cleanup=True,
                    #csv_filename=cut_face_filename,
                    csv_filename=None,
                    #plane_bdf_filename=None)
                    plane_bdf_filename=plane_bdf_filename,
                    plane_bdf_offset=dy)
            except RuntimeError:
                # incorrect ivalues=[0, 1, 2]; dy=771. for CRM
                continue
            unused_unique_geometry_array, unused_unique_results_array, rods = out

            if not os.path.exists(plane_bdf_filename):
                break
            plane_bdf_filenames.append(plane_bdf_filename)
            # eid, nid, inid1, inid2
            #print(unique_geometry_array)
            #moi_filename = 'amoi_%i.bdf' % i
            moi_filename = None
            out = calculate_area_moi(model,
                                     rods,
                                     normal_plane,
                                     thetas,
                                     moi_filename=moi_filename)
            #print(out)
            Ai, Ii, EIi, avg_centroidi = out
            y.append(dy)
            A.append(Ai)
            I.append(Ii)
            EI.append(EIi)
            avg_centroid.append(avg_centroidi)
            #break

        with open('thetas.csv', 'w') as csv_filename:
            csv_filename.write('# eid(%i),theta,Ex,Ey,Gxy\n')
            for eid, (theta, Ex, Ey, Gxy) in sorted(thetas.items()):
                csv_filename.write('%i,%f,%f,%f,%f\n' %
                                   (eid, theta, Ex, Ey, Gxy))

        y = np.array(y, dtype='float64')
        A = np.array(A, dtype='float64')
        I = np.array(I, dtype='float64')
        EI = np.array(EI, dtype='float64')
        avg_centroid = np.array(avg_centroid, dtype='float64')

        inid = 1
        beam_model = BDF(debug=False)
        avg_centroid[:, 1] = y

        # wrong
        mid = 1
        E = 3.0e7
        G = None
        nu = 0.3
        model.add_mat1(mid, E, G, nu, rho=0.1)

        Ix = I[:, 0]
        Iy = I[:, 1]
        Ixy = I[:, 2]
        J = Ix + Iy
        #i1, i2, i12 = Ix, Iy, Ixy
        for inid, xyz in enumerate(avg_centroid):
            beam_model.add_grid(inid + 1, xyz)
        for eid in range(1, len(A)):
            pid = eid
            nids = [eid, eid + 1]
            x = [1., 0., 0.]
            g0 = None
            beam_model.add_cbeam(eid,
                                 pid,
                                 nids,
                                 x,
                                 g0,
                                 offt='GGG',
                                 bit=None,
                                 pa=0,
                                 pb=0,
                                 wa=None,
                                 wb=None,
                                 sa=0,
                                 sb=0,
                                 comment='')

            # j = i1 + i2
            so = ['YES', 'YES']
            xxb = [0., 1.]
            area = [A[eid - 1], A[eid]]
            i1 = [Ix[eid - 1], Ix[eid]]
            i2 = [Iy[eid - 1], Iy[eid]]
            i12 = [Ixy[eid - 1], Ixy[eid]]
            j = [J[eid - 1], J[eid]]
            beam_model.add_pbeam(pid,
                                 mid,
                                 xxb,
                                 so,
                                 area,
                                 i1,
                                 i2,
                                 i12,
                                 j,
                                 nsm=None,
                                 c1=None,
                                 c2=None,
                                 d1=None,
                                 d2=None,
                                 e1=None,
                                 e2=None,
                                 f1=None,
                                 f2=None,
                                 k1=1.,
                                 k2=1.,
                                 s1=0.,
                                 s2=0.,
                                 nsia=0.,
                                 nsib=None,
                                 cwa=0.,
                                 cwb=None,
                                 m1a=0.,
                                 m2a=None,
                                 m1b=0.,
                                 m2b=None,
                                 n1a=0.,
                                 n2a=None,
                                 n1b=0.,
                                 n2b=None,
                                 comment='')
        beam_model.write_bdf('equivalent_beam_model.bdf')

        X = np.vstack([y, A]).T
        Y = np.hstack([X, I, EI, avg_centroid])
        header = 'y, A, Ix, Iz, Ixz, Ex*Ix, Ex*Iz, Ex*Ixz, xcentroid, ycentroid, zcentroid'
        np.savetxt('cut_data_vs_span.csv', Y, header=header, delimiter=',')

        show = True
        #show = False
        if IS_MATPLOTLIB:
            plot_inertia(y, A, I, EI, avg_centroid, show=show)
Exemple #2
0
def cut_and_plot_moi(bdf_filename: str, normal_plane: np.ndarray, log: SimpleLogger,
                     dys: List[float],
                     coords: List[CORD2R],
                     ytol: float=2.0,
                     dirname: str='',
                     plot: bool=True, show: bool=False) -> Tuple[Any, Any, Any, Any, Any]: # y, A, I, EI, avg_centroid
    model = read_bdf(bdf_filename, log=log)
    model2 = read_bdf(bdf_filename, log=log)

    # initialize theta
    thetas = {}
    for eid in model.elements:
        #  theta, Ex, Ey, Gxy
        thetas[eid] = (0., 0., 0., 0.)

    #p1 = np.array([466.78845, 735.9053, 0.0])
    #p2 = np.array([624.91345, 639.68896, -0.99763656])
    #dx = p2 - p1
    nodal_result = None
    plane_bdf_filenames = []
    y = []
    A = []
    I = []
    J = []
    EI = []
    GJ = []
    avg_centroid = []

    for i, dy, coord in zip(count(), dys, coords):
        model.coords[1] = coord
        plane_bdf_filename = os.path.join(dirname, f'plane_face_{i:d}.bdf')
        cut_face_filename = os.path.join(dirname, f'cut_face_{i:d}.csv')
        if os.path.exists(cut_face_filename):
            os.remove(cut_face_filename)
        try:
            out = cut_face_model_by_coord(
                model2, coord, ytol,
                nodal_result, plane_atol=1e-5, skip_cleanup=True,
                #csv_filename=cut_face_filename,
                csv_filename=None,
                #plane_bdf_filename=None)
                plane_bdf_filename=plane_bdf_filename, plane_bdf_offset=dy)
        except RuntimeError:
            # incorrect ivalues=[0, 1, 2]; dy=771. for CRM
            continue
        unused_unique_geometry_array, unused_unique_results_array, rods = out

        if not os.path.exists(plane_bdf_filename):
            break
        plane_bdf_filenames.append(plane_bdf_filename)
        # eid, nid, inid1, inid2
        #print(unique_geometry_array)
        #moi_filename = 'amoi_%i.bdf' % i
        moi_filename = None
        out = calculate_area_moi(model, rods, normal_plane, thetas, moi_filename=moi_filename)

        #print(out)
        Ai, Ii, EIi, avg_centroidi = out
        #Ai, Ii, Ji, EIi, GJi, avg_centroidi = out
        Ji = GJi = 1.0
        y.append(dy)
        A.append(Ai)
        I.append(Ii)
        J.append(Ji)
        EI.append(EIi)
        GJ.append(GJi)
        avg_centroid.append(avg_centroidi)
        #break

    thetas_csv_filename = os.path.join(dirname, 'thetas.csv')

    with open(thetas_csv_filename, 'w') as csv_filename:
        csv_filename.write('# eid(%i),theta,Ex,Ey,Gxy\n')
        for eid, (theta, Ex, Ey, Gxy) in sorted(thetas.items()):
            csv_filename.write('%i,%f,%f,%f,%f\n' % (eid, theta, Ex, Ey, Gxy))

    y = np.array(y, dtype='float64')
    A = np.array(A, dtype='float64')
    I = np.array(I, dtype='float64')
    J = np.array(J, dtype='float64')
    EI = np.array(EI, dtype='float64')
    GJ = np.array(GJ, dtype='float64')
    avg_centroid = np.array(avg_centroid, dtype='float64')

    inid = 1
    beam_model = BDF(debug=False)
    avg_centroid[:, 1] = y

    # wrong
    mid = 1
    E = 3.0e7
    G = None
    nu = 0.3
    model.add_mat1(mid, E, G, nu, rho=0.1)

    #   0    1    2    3    4    5
    # [Ixx, Iyy, Izz, Ixy, Iyz, Ixz]
    Ix = I[:, 0]
    Iy = I[:, 1]
    Iz = I[:, 2]
    Ixz = I[:, 5]
    J = Ix + Iz
    #i1, i2, i12 = Ix, Iy, Ixy
    for inid, xyz in enumerate(avg_centroid):
        beam_model.add_grid(inid+1, xyz)
    for eid in range(1, len(A)):
        pid = eid
        nids = [eid, eid + 1]
        x = [1., 0., 0.]
        g0 = None
        beam_model.add_cbeam(eid, pid, nids, x, g0, offt='GGG', bit=None,
                             pa=0, pb=0, wa=None, wb=None, sa=0, sb=0, comment='')

        # j = i1 + i2
        so = ['YES', 'YES']
        xxb = [0., 1.]
        area = [A[eid-1], A[eid]]
        i1 = [Ix[eid-1], Ix[eid]]
        i2 = [Iz[eid-1], Iz[eid]]
        i12 = [Ixz[eid-1], Ixz[eid]]
        j = [J[eid-1], J[eid]]
        beam_model.add_pbeam(pid, mid, xxb, so, area, i1, i2, i12, j, nsm=None,
                             c1=None, c2=None, d1=None, d2=None, e1=None, e2=None, f1=None, f2=None,
                             k1=1., k2=1., s1=0., s2=0., nsia=0., nsib=None, cwa=0., cwb=None,
                             m1a=0., m2a=0., m1b=None, m2b=None,
                             n1a=0., n2a=0., n1b=None, n2b=None,
                             comment='')

    beam_model_bdf_filename = os.path.join(dirname, 'equivalent_beam_model.bdf')
    beam_model.write_bdf(beam_model_bdf_filename)

    X = np.vstack([y, A]).T
    Y = np.hstack([X, I, EI, avg_centroid])
    header = 'y, A, Ix, Iz, Ixz, Ex*Ix, Ex*Iz, Ex*Ixz, xcentroid, ycentroid, zcentroid'
    cut_data_span_filename = os.path.join(dirname, 'cut_data_vs_span.csv')
    np.savetxt(cut_data_span_filename, Y, header=header, delimiter=',')

    if IS_MATPLOTLIB and (plot or show):
        plot_inertia(y, A, I, J, EI, GJ, avg_centroid, show=show, dirname=dirname)
    else:
        plane_bdf_filenames = []
    return y, A, I, J, EI, GJ, avg_centroid, plane_bdf_filenames