wawi 0.0.1__py3-none-any.whl → 0.0.3__py3-none-any.whl

wawi/plot.py

@@ -0,0 +1,569 @@
+import numpy as np
+import matplotlib.pyplot as plt
+from matplotlib.animation import FuncAnimation 
+from wawi.wave import dispersion_relation_scalar as get_kappa
+from scipy.ndimage import rotate, shift
+from matplotlib import transforms
+from scipy.interpolate import RectBivariateSpline, interp1d
+
+
+def plot_ads(ad_dict, v, terms='stiffness', num=None, test_v=dict(), test_ad=dict(), zasso_type=False, ranges=None):
+    # v: v or K
+    if terms is 'stiffness':
+        terms = [['P4', 'P6', 'P3'], ['H6', 'H4', 'H3'], ['A6', 'A4', 'A3']]
+    elif terms is 'damping':
+        terms = [['P1', 'P5', 'P2'], ['H5', 'H1', 'H2'], ['A5', 'A1', 'A2']]
+
+    # Create exponent defs for K_normalized plotting
+    K_exp = dict()
+    stiffness_terms = ['P4', 'P6', 'P3', 'H6', 'H4', 'H3', 'A6', 'A4', 'A3']
+    damping_terms = ['P1', 'P5', 'P2', 'H5', 'H1', 'H2', 'A5', 'A1', 'A2']
+    K_exp.update(zip(stiffness_terms, [1]*len(stiffness_terms)))
+    K_exp.update(zip(damping_terms, [2]*len(damping_terms)))
+    
+    K_label_add = dict()
+    K_label_add.update(zip(stiffness_terms, ['$K$']*len(stiffness_terms)))
+    K_label_add.update(zip(damping_terms, ['$K^2$']*len(damping_terms)))
+        
+    fig, ax = plt.subplots(nrows=len(terms[0]), ncols=len(terms), num=num, sharex=True)
+    
+    for row_ix, row in enumerate(terms):
+        for col_ix, term in enumerate(row):
+            axi = ax[row_ix, col_ix]
+            
+            if term in ad_dict:
+                if ranges is not None and term in ranges:
+                    ad_valid = ad_dict[term](v)*1
+                    
+                    v_min, v_max = ranges[term]
+                    
+                    ad_valid[v<v_min] = ad_dict[term](v_min)
+                    ad_valid[v>v_max] = ad_dict[term](v_max)
+                else:
+                    ad_valid = ad_dict[term](v)
+                
+                axi.plot(v, ad_valid, label='Fit')
+                
+            label = zasso_type*K_label_add[term]+('$' + term[0] + '_' + term[1] + '^*' + '$')
+            axi.set_ylabel(label)
+            axi.grid('on')
+            
+            if term in test_v:
+                if zasso_type:
+                    vK = 1/test_v[term]
+                    factor = vK**K_exp[term]
+                else:
+                    vK = test_v[term]
+                    factor = 1.0
+                    
+                axi.plot(vK, test_ad[term]*factor, '.k', label='Test')
+    
+    for col_ix in range(len(terms)):
+        if zasso_type:
+            ax[-1, col_ix].set_xlabel('$K$')
+        else:
+            ax[-1, col_ix].set_xlabel('$V/(B\cdot \omega)$')
+    
+    fig.tight_layout()
+    return fig, ax
+
+def save_plot(pl, path, w=None, h=None):
+    ws = pl.window_size
+    if w is not None and h is None:
+        w = int(np.round(w))
+        h = int(np.round(ws[1] * w/ws[0]))
+    elif h is not None and w is None:
+        h = int(np.round(h))
+        w = int(np.round(ws[0] * h/ws[1]))
+    elif h is None and w is None:
+        w,h = ws
+    else:
+        w = int(np.round(w))
+        h = int(np.round(h))
+
+    pl.screenshot(path, window_size=[w,h], return_img=False)
+
+def plot_dir_and_crests(theta0, Tp, arrow_length=100, origin=np.array([0,0]), 
+                        ax=None, n_repeats=2, crest_length=1000,
+                        alpha_crests=0.2, arrow_options={}):
+    arr_opts = {'head_width': 4, 'width':2, 'edgecolor':'none'}
+    arr_opts.update(**arrow_options)
+    
+    if ax is None:
+        ax = plt.gca()
+        
+    # Plot wave angle and crests
+    v = np.array([np.cos(theta0*np.pi/180), np.sin(theta0*np.pi/180)])
+    v_norm = np.array([-np.sin(theta0*np.pi/180), np.cos(theta0*np.pi/180)])
+    wave_length = 2*np.pi/get_kappa(2*np.pi/Tp, U=0.0)
+
+    plt.arrow(origin[0],origin[1], arrow_length*v[0], arrow_length*v[1], **arr_opts)
+    plt.text(origin[0], origin[1], f'$\\theta_0$ = {theta0}$^o$\n $T_p$={Tp} s\n $\lambda=${wave_length:.0f} m')
+
+    dv = v*wave_length
+    for n in range(n_repeats):
+        p1 = origin-v_norm*crest_length/2
+        p2 = origin+v_norm*crest_length/2
+        pts = np.vstack([p1,p2])
+        
+        ax.plot(pts[:,0], pts[:,1], alpha=alpha_crests, color='black', zorder=0)
+        origin = origin + dv
+
+    return ax
+
+def rotate_image_about_pivot(Z, x, y, angle, x0=0, y0=0):
+    xc = np.mean(x)
+    yc = np.mean(y)
+    
+    pixel_x = interp1d(x, np.arange(len(x)), fill_value='extrapolate') 
+    pixel_y = interp1d(y, np.arange(len(y)), fill_value='extrapolate')   
+
+    ds_x = pixel_x(x0) - pixel_x(xc) # sample shift x
+    ds_y = pixel_y(y0) - pixel_y(yc) # sample shift y
+    ds = np.array([ds_x, ds_y])*0
+    
+    T = np.array([[np.cos(angle*np.pi/180), np.sin(angle*np.pi/180)],
+                  [-np.sin(angle*np.pi/180), np.cos(angle*np.pi/180)]])
+                  
+    ds_rot = T @ ds
+    
+    return shift(rotate(shift(Z, ds[::-1]), angle), -ds_rot[::-1])
+
+def combine_eta(eta_fine, eta_course, x_fine, y_fine, x_course, y_course, x=None, y=None):
+    dx_fine = x_fine[1]-x_fine[0]
+    dy_fine = y_fine[1]-y_fine[0]
+    
+    if x is None:
+        x = np.arange(np.min(x_course), np.max(x_course), dx_fine)
+    
+    if y is None:
+        y = np.arange(np.min(y_course), np.max(y_course), dy_fine)
+    
+    eta_course_i = RectBivariateSpline(y_course,x_course,eta_course)
+    eta_fine_i = RectBivariateSpline(y_fine,x_fine,eta_fine)
+    
+    eta_combined = eta_course_i(y,x)
+    sel = np.ix_(
+        (y >= np.min(y_fine)) & (y <= np.max(y_fine)),
+        (x >= np.min(x_fine)) & (x <= np.max(x_fine)))
+    
+    eta_combined[sel] = eta_fine_i(y[(y >= np.min(y_fine)) & (y <= np.max(y_fine))],
+                                   x[(x >= np.min(x_fine)) & (x <= np.max(x_fine))])
+    
+    return eta_combined, x, y
+
+
+def animate_surface(eta, x, y, t, filename=None, fps=None, 
+                    speed_ratio=1.0, figsize=None, writer='ffmpeg', 
+                    ax=None, surface=None):
+    
+        
+    if surface is None:
+        if ax is None:
+            fig, ax = plt.subplots(figsize=figsize)   
+        else:
+            fig = fig.get_figure()
+        surface, __ = plot_surface(eta[:,:,0], x, y, ax=ax, colorbar=False, labels=False)
+    else:
+        fig = surface.get_figure()
+        ax = fig.axes[0]
+    
+    def animate(i): 
+        surface.set_data(eta[:,:,i])
+        return surface, 
+    
+    frames = np.arange(len(t))
+    dt = t[1]-t[0]
+    fs = 1/dt
+    
+    if filename is None:
+        repeat = True
+    else:
+        repeat = False
+    
+    if fps is None:
+        fps = fs*speed_ratio
+        
+    interval = 1/fps*1000
+    
+    anim = FuncAnimation(fig, animate, 
+                         frames=frames, interval=interval, blit=True, repeat=repeat) 
+    
+    if filename is not None:
+        anim.save(filename, writer=writer, fps=fps)
+    else:
+        plt.show()
+
+def xy_to_latlon(latlon0, coors, rot=0):
+    import cartopy.crs as ccrs, cartopy.geodesic as cgds
+    dist = np.linalg.norm(coors, axis=1)
+    azi = np.arctan2(coors[:,0], coors[:,1])*180/np.pi - rot # atan2(x,y) to get azimuth (relative to N-vector)
+    geodesic = cgds.Geodesic()
+    
+    return np.vstack(geodesic.direct(np.tile(latlon0, [len(dist), 1]), azi, dist))[:,:2]
+
+
+def plot_surface_in_map(eta, x, y, eta_geo0, extent, 
+                        eta_scatter=None, 
+                        wms_url='https://openwms.statkart.no/skwms1/wms.terrengmodell?request=GetCapabilities&service=WMS', 
+                        wms_layers=['relieff'], ax=None,
+                        cm='Blues_r', colorbar=True, figsize=None, eta_rot=0, labels=False):
+    
+    import cartopy.crs as ccrs, cartopy.geodesic as cgds
+    
+    proj = 'Mercator'
+    proj = getattr(ccrs, proj)()
+    
+
+    if ax is None:
+        ax = plt.axes(projection=proj)
+
+    
+    # Plot scatter
+    if eta_scatter is None:
+        scatter = None
+    else:
+        eta_scatter = xy_to_latlon(eta_geo0, eta_scatter, rot=eta_rot)
+        scatter = ax.scatter(eta_scatter[:,0], eta_scatter[:,1], c='black', s=6, transform=ccrs.PlateCarree())
+    
+    # Plot eta
+    if eta is not None:
+        eta_max = np.max(np.abs(eta))
+        corners = np.array([[np.min(x), np.min(y)],
+                    [np.min(x), np.max(y)],
+                    [np.max(x), np.max(y)],
+                    [np.max(x), np.min(y)]])
+        
+        corners_latlon = xy_to_latlon(eta_geo0, corners, rot=eta_rot)
+
+        extent_merc = np.vstack(proj.transform_points(ccrs.PlateCarree(), np.array([extent[0], extent[2]]), np.array([extent[1], extent[3]])))[:,:2]     
+        extent_merc = [np.min(extent_merc[:,0]), np.max(extent_merc[:,0]), 
+                       np.min(extent_merc[:,1]), np.max(extent_merc[:,1])]
+        
+        ax.imshow(np.zeros([2,2]), cmap=cm, origin='lower', interpolation='none', 
+                    vmin=-eta_max, vmax=eta_max, extent=extent_merc)
+        
+        corners_latlon_new = np.vstack(proj.transform_points(ccrs.PlateCarree(), *corners_latlon.T))[:,:2]
+        eta_extent_new = [np.min(corners_latlon_new[:,0]), np.max(corners_latlon_new[:,0]), 
+                  np.min(corners_latlon_new[:,1]), np.max(corners_latlon_new[:,1])]
+        
+        eta_rotated = rotate(eta, -eta_rot)
+        
+        surface = ax.imshow(eta_rotated, cmap=cm, origin='lower', 
+                            interpolation='none', extent=eta_extent_new, 
+                            vmin=-eta_max, vmax=eta_max)   
+        
+        if colorbar:
+            plt.colorbar(surface)        
+            
+    else:
+        surface = None
+    if labels:
+        ax.gridlines(color='lightgrey', linestyle='-', draw_labels=True)
+        
+    ax.add_wms(wms_url, layers=wms_layers, extent=extent)
+    ax.set_extent(extent)
+
+    return ax, scatter, surface
+
+
+def plot_surface(eta, x, y, ax=None, 
+                 cm='Blues_r', colorbar=True, 
+                 labels=True, figsize=None, interpolation='none'): 
+
+    if ax is None:
+        fig, ax = plt.subplots(figsize=figsize)
+
+    dx = (x[1]-x[0])/2.
+    dy = (y[1]-y[0])/2.
+    extent = [x[0]-dx, x[-1]+dx, y[0]-dy, y[-1]+dy]
+
+    eta_max = np.max(np.abs(eta))   
+    surface_plot = ax.imshow(eta, cmap=cm, extent=extent, origin='lower', vmin=-eta_max, vmax=eta_max, interpolation=interpolation)
+    
+    if labels:
+        ax.set_xlabel('x [m]')
+        ax.set_ylabel('y [m]')
+    else:
+        ax.set_xticks([])
+        ax.set_yticks([])
+    
+    if colorbar:
+        cb = plt.colorbar(surface_plot)
+    
+    return surface_plot, ax
+
+
+def set_axes_equal(ax: plt.Axes):
+    import matplotlib.pyplot as plt
+    """Set 3D plot axes to equal scale.
+
+    Make axes of 3D plot have equal scale so that spheres appear as
+    spheres and cubes as cubes.  Required since `ax.axis('equal')`
+    and `ax.set_aspect('equal')` don't work on 3D.
+    """
+    limits = np.array([
+        ax.get_xlim3d(),
+        ax.get_ylim3d(),
+        ax.get_zlim3d(),
+    ])
+    origin = np.mean(limits, axis=1)
+    radius = 0.5 * np.max(np.abs(limits[:, 1] - limits[:, 0]))
+    _set_axes_radius(ax, origin, radius)
+
+def _set_axes_radius(ax, origin, radius):
+    x, y, z = origin
+    ax.set_xlim3d([x - radius, x + radius])
+    ax.set_ylim3d([y - radius, y + radius])
+    ax.set_zlim3d([z - radius, z + radius])
+
+def equal_3d(ax=plt.gca()):
+    x_lims = np.array(ax.get_xlim())
+    y_lims = np.array(ax.get_ylim())
+    z_lims = np.array(ax.get_zlim())
+
+    x_range = np.diff(x_lims)
+    y_range = np.diff(y_lims)
+    z_range = np.diff(z_lims)
+
+    max_range = np.max([x_range,y_range,z_range])/2
+
+    ax.set_xlim(np.mean(x_lims) - max_range, np.mean(x_lims) + max_range)
+    ax.set_ylim(np.mean(y_lims) - max_range, np.mean(y_lims) + max_range)
+    ax.set_zlim(np.mean(z_lims) - max_range, np.mean(z_lims) + max_range)
+    # ax.set_aspect(1)
+    
+    return ax
+
+def plot_transformation_mats(x,y,z,T,figno=None, ax=None, scaling='auto'):
+    
+    if ax==None:
+        fig = plt.figure(figno)
+        ax = fig.add_subplot(111, projection='3d')
+
+    ax.scatter(x,y,z,'.k')
+
+    if scaling=='auto':
+        xr = max(x)-min(x)
+        yr = max(y)-min(y)
+        zr = max(z)-min(z)
+        r = np.sqrt(xr**2+yr**2+zr**2)
+        scaling = 0.005*r
+
+    compcolors = ['tab:red', 'tab:blue', 'tab:green']
+    h = [None]*3
+    for ix, Ti in enumerate(T):
+        xi = x[ix]
+        yi = y[ix]
+        zi = z[ix]
+        
+        for comp in range(0,3):
+            xunit = [xi, xi+Ti[comp,0]*scaling]
+            yunit = [yi, yi+Ti[comp,1]*scaling]
+            zunit = [zi, zi+Ti[comp,2]*scaling]
+
+            h[comp] = plt.plot(xs=xunit,ys=yunit,zs=zunit, color=compcolors[comp])[0]
+
+    plt.legend(h,['x', 'y', 'z'])
+    ax.set_xlabel('x')
+    ax.set_ylabel('y')
+    ax.set_zlabel('z')
+
+    equal_3d(ax)
+    return ax,h
+
+
+def plot_elements(element_matrix, node_matrix, chosen_nodes_ix=[], disp=None, node_labels=False, element_labels=False, plot_nodes=True, plot_elements=True, ax=None, fig=None, element_settings={}, node_settings={}, node_label_settings={}, chosen_node_settings={}, disp_settings={}, element_label_settings={}, three_d=True):
+    e_dict = {'color': 'LimeGreen', 'alpha': 1}
+    e_dict.update(**element_settings)
+
+    n_dict = {'color':'Black', 'linestyle':'', 'marker':'.', 'markersize':4, 'alpha':0.8}
+    n_dict.update(**node_settings)
+       
+    n_chosen_dict = {'color':'GreenYellow', 'linestyle':'', 'marker':'o', 'markersize':8, 'alpha':1, 'markeredgecolor':'dimgray'}
+    n_chosen_dict.update(**chosen_node_settings)
+    
+    disp_dict = {'color':'IndianRed', 'alpha':1}
+    disp_dict.update(**disp_settings)
+
+    l_nodes_dict = {'color':'Black', 'fontsize': 8, 'fontweight':'normal'}
+    l_nodes_dict.update(**node_label_settings)
+    
+    l_elements_dict = {'color':'LimeGreen', 'fontsize': 8, 'fontweight':'bold', 'style':'italic'}
+    l_elements_dict.update(**element_label_settings)
+    
+    if ax is None and fig is None:
+        fig = plt.figure()
+        
+    if ax == None and three_d:
+        ax = fig.gca(projection='3d')
+    elif ax == None:
+        ax = fig.gca()
+    elif three_d:
+        1
+        # ax.set(projection='3d')  #mangler funksjonalitet her...
+    
+    element_handles = [None]*len(element_matrix[:,0])
+
+    if plot_elements:
+        for element_ix, __ in enumerate(element_matrix[:,0]):
+            node1 = element_matrix[element_ix, 1]
+            node2 = element_matrix[element_ix, 2]
+            nodeix1 = np.where(node_matrix[:,0]==node1)[0]
+            nodeix2 = np.where(node_matrix[:,0]==node2)[0]
+            x01 = node_matrix[nodeix1,1:4]
+            x02 = node_matrix[nodeix2,1:4]
+            x0 = np.vstack([x01,x02])
+
+            if three_d:
+                element_handles[element_ix] = ax.plot(xs=x0[:,0], ys=x0[:,1], zs=x0[:,2], **e_dict)
+            else:
+                element_handles[element_ix] = ax.plot(x0[:,0], x0[:,1], **e_dict)
+  
+            if element_labels:
+                xmean = np.mean(x0, axis=0)
+                if three_d:
+                    ax.text(xmean[0],xmean[1],xmean[2],'%i' % element_matrix[element_ix,0], **l_elements_dict)
+                else:
+                    ax.text(xmean[0],xmean[1],s='%i' % element_matrix[element_ix,0], **l_elements_dict)
+
+            if disp is not None:
+                disp_node1 = disp[nodeix1[0]*6:(nodeix1[0]*6+6)]
+                disp_node2 = disp[nodeix2[0]*6:(nodeix2[0]*6+6)]
+                x1 = x01+disp_node1[0:3]
+                x2 = x02+disp_node2[0:3]
+                x = np.vstack([x1,x2])
+                
+                if three_d:
+                    ax.plot(xs=x[:,0], ys=x[:,1], zs=x[:,2], **disp_dict)
+                else:
+                    ax.plot(x[:,0], x[:,1], **disp_dict)
+
+    if plot_nodes:
+        if three_d:
+            ax.plot(xs=node_matrix[:, 1], ys=node_matrix[:, 2], zs=node_matrix[:, 3], **n_dict)
+        else:
+           ax.plot(node_matrix[:, 1], node_matrix[:, 2], **n_dict)
+        
+        if chosen_nodes_ix != []:
+            if three_d:
+                ax.plot(xs=node_matrix[chosen_nodes_ix, 1], ys=node_matrix[chosen_nodes_ix, 2], zs=node_matrix[chosen_nodes_ix, 3], **n_chosen_dict)
+            else:
+               ax.plot(node_matrix[chosen_nodes_ix, 1], node_matrix[chosen_nodes_ix, 2], **n_chosen_dict)
+        
+    if node_labels:
+        if three_d:
+            for node_ix in range(0, np.shape(node_matrix)[0]):
+                ax.text(node_matrix[node_ix, 1], node_matrix[node_ix, 2], node_matrix[node_ix, 3], '%i' % node_matrix[node_ix, 0], **l_nodes_dict)
+        else:
+            for node_ix in range(0, np.shape(node_matrix)[0]):
+                ax.text(node_matrix[node_ix, 1], node_matrix[node_ix, 2], '%i' % node_matrix[node_ix, 0], **l_nodes_dict)
+        
+    if three_d:
+        equal_3d(ax)
+    else:
+        ax.set_aspect('equal', adjustable='box')
+    
+    ax.grid('off')
+    return ax, element_handles
+
+
+def plot_2d(S2d, x1, x2, ax=None, levels=80, discrete=False, **kwargs):
+    if ax is None:
+        ax = plt.gca()
+        
+    X, Y = np.meshgrid(x1, x2)
+    if discrete:
+        contour = ax.pcolormesh(x1,x2,S2d, **kwargs)
+    else:        
+        contour = ax.contourf(X, Y, S2d.T, levels=levels, **kwargs)
+    return contour
+
+
+def plot_S2d(S, omega, theta, D=None, omega_range=None, theta_range=None):
+
+    if theta_range is None:
+        theta_range = [np.min(theta), np.max(theta)]
+    
+    if omega_range is None:
+        omega_range = [0, np.max(omega)]
+
+    X, Y = np.meshgrid(omega, theta)
+
+    if D is None:
+        SD = S*1
+    else:
+        SD = S[:,np.newaxis] @ D[np.newaxis,:]
+    
+    SD[np.isnan(SD)] = 0
+    
+    plt.figure(2).clf()
+    fig = plt.figure(num=2, constrained_layout=True)
+    
+    if D is not None:
+        widths = [2, 0.7]
+        heights = [0.7, 2.0, 0.7]
+        spec = fig.add_gridspec(ncols=2, nrows=3, width_ratios=widths,
+                              height_ratios=heights, wspace=.1, hspace=.1)
+    else:
+        widths = [2]
+        heights = [2.0, 0.7]
+        spec = fig.add_gridspec(ncols=1, nrows=2, width_ratios=widths,
+                              height_ratios=heights, wspace=.1, hspace=.1)        
+    
+    if D is not None:
+        ax = [None]*3
+        ax[0] = fig.add_subplot(spec[1,0])   # SD
+        ax[1] = fig.add_subplot(spec[0,0])   # S
+        ax[2] = fig.add_subplot(spec[1,1])   # D
+        ax[1].set_yticklabels('')
+        ax[2].set_xticklabels('')
+        ax[2].set_yticklabels('')
+        cbar_ax = fig.add_subplot(spec[2,0])
+    else:
+        ax = [fig.add_subplot(spec[0,0])]
+        cbar_ax = fig.add_subplot(spec[1,0])
+
+    cbar_ax.axis('off')
+
+    # Contour plot
+    contour = ax[0].contourf(X, Y, SD.T)
+    ax[0].set_ylim(theta_range)
+    ax[0].set_xlim(omega_range)
+    ax[0].set_ylabel(r'$\theta$ [rad]')
+    ax[0].set_xlabel(r'$\omega$ [rad/s]')
+   
+    if D is not None:
+        # S line plot
+        ax[1].plot(omega, S)
+        ax[1].set_ylim(bottom=0)
+        ax[1].set_xlim(omega_range)
+        ax[1].set_xticklabels('')
+        ax[1].set_yticks([])
+        
+        # D line plot
+
+        ax[2].plot(D, theta)
+        ax[2].set_ylim(theta_range)
+        ax[2].set_xlim(left=0)
+        ax[2].set_xticks([])
+    
+    # cbar_ax.axis('off')
+    fig.colorbar(contour, ax=cbar_ax, orientation="horizontal", aspect=25, shrink=1.0)
+
+    ax[0].spines['top'].set_visible(False)
+    ax[0].spines['right'].set_visible(False)
+
+    if D is not None:
+        ax[1].spines['top'].set_visible(False)
+        ax[1].spines['right'].set_visible(False)
+        ax[1].set_ylabel(r'$S_\eta(\omega)$')
+        
+        ax[2].spines['bottom'].set_visible(False)
+        ax[2].spines['right'].set_visible(False)
+        ax[2].set_xlabel(r'$D(\theta)$',rotation=-90)
+        ax[2].xaxis.set_label_position('top') 
+    
+        fig.subplots_adjust(top=0.97, bottom=0.08, left=0.16, right=0.97)
+    
+    return fig

wawi/prob.py

@@ -0,0 +1,9 @@
+import numpy as np
+
+def gumbel_log(umax):
+    umax_ordered = np.sort(umax)[::-1]
+    N_stat = len(umax)
+    F = 1-np.arange(1, N_stat+1)/(N_stat+1)
+    loglogF = -np.log(-np.log(F))
+
+    return umax_ordered, loglogF

wawi/random.py

@@ -0,0 +1,38 @@
+import numpy as np
+
+def zero_crossing_period(S, omega):
+    return 2*np.pi*np.sqrt(np.trapz(S, x=omega)/np.trapz(omega**2*S, x=omega))
+
+def stoch_mom(S, omega, n=0):
+    return np.trapz(S*omega**n, x=omega)
+
+def m0(S, omega):
+    return stoch_mom(S, omega, n=0)
+
+def m2(S, omega):
+    return stoch_mom(S, omega, n=2)
+
+def v0_from_spectrum(S, omega):
+    return 1/(2*np.pi) * np.sqrt(m2(S, omega)/m0(S, omega))
+
+def v0(m0,m2):
+    return 1/(2*np.pi) * np.sqrt(m2/m0)
+
+def peakfactor(T, v0):
+    c = np.sqrt(2*np.log(v0*T))
+    kp = c + np.euler_gamma/c
+    return kp
+
+def expmax(T, v0, std):
+    return peakfactor(T,v0)*std
+
+def expmax_from_spectrum(S, omega, T):
+    m0_val = m0(S, omega)
+    std = np.sqrt(m0_val)
+
+    v0_val = v0(m0_val, m2(S, omega))
+
+    return expmax(T, v0_val, std)
+
+def peakfactor_from_spectrum(S, omega, T):
+    return peakfactor(T, v0(S, omega))

wawi/signal.py

@@ -0,0 +1,45 @@
+from .wave import jonswap
+import numpy as np
+from scipy.optimize import curve_fit
+
+def fit_jonswap(S, w, sigma=[0.07, 0.09], initial_values=None):
+    if initial_values is None:
+        initial_values = {}
+        
+    Hs0 = np.sqrt(np.trapz(S, w))*4
+
+    p0 = {'Hs': Hs0, 'Tp': 1, 'gamma': 1}     #standard values of curve_fit are 1
+    p0.update(**initial_values)
+
+    if w[0]==0:
+        w = w[1:]
+        S = S[1:]
+
+    fun = lambda om, Hs, Tp, gamma: jonswap(Hs, Tp, gamma, sigma=sigma)(om)
+    popt,__ = curve_fit(fun, w, S)
+    out = dict(Hs=popt[0], Tp=popt[1], gamma=popt[2], p0=[p0['Hs'], p0['Tp'], p0['gamma']])
+
+    return out
+    
+def onesided_to_twosided(omega, S, axis=-1):
+    S2 = 0.5*np.concatenate([np.flip(S, axis=axis), S], axis=axis)
+    omega2 = np.hstack([np.flip(-omega), omega])
+    
+    return omega2, S2
+
+
+def twosided_to_onesided(omega, S):
+    n_samples = len(omega)    
+    return omega[:n_samples//2], S[:,:,:n_samples//2]
+
+
+def ramp_up(Nramp, Ntot):
+    t_scale = np.ones(Ntot)
+    t_scale[:Nramp] = np.linspace(0, 1, Nramp)
+    return t_scale
+
+def ramp_up_t(t, t0):
+    Nramp = np.sum(t<t0)
+    Ntot = len(t)
+    
+    return ramp_up(Nramp, Ntot)