From 4a1868878f92ad97700bb82151bd309abdf550e1 Mon Sep 17 00:00:00 2001
From: weisse02 <andreas.weissenbrunner@ptb.de>
Date: Wed, 22 Feb 2023 17:42:43 +0100
Subject: [PATCH] Changed the regGridinterpolators get_profile works
---
Flow_class.py | 109 +++++++++++++++-----------------------------------
1 file changed, 33 insertions(+), 76 deletions(-)
diff --git a/Flow_class.py b/Flow_class.py
index 091ddbd..cbe9045 100644
--- a/Flow_class.py
+++ b/Flow_class.py
@@ -117,10 +117,9 @@ def loadPODres(case):
#
case_ind = [0, 2121, 26321, 44592]
# distinguish between the cases 0, 1, 2 for SE,DE,DSE
- data["coeffs"] = data["coeffs"][case_ind[ci-1]:case_ind[ci]]
-
-
-
+ coeffs = data["coeffs"][case_ind[ci-1]:case_ind[ci]]
+ modes = data["modes"]
+ umean = data["umean"]
# if case == "SingleElbow":
# # the POD res files are in npz-format
# ux = np.load(data_path+"/Ux_POD_SE.npz")
@@ -137,13 +136,12 @@ def loadPODres(case):
# uy = np.load(data_path+"/Uy_POD_DSE.npz")
# uz = np.load(data_path+"/Udiff_POD_DSE.npz")
#
- return data
-
-
-
-
-
-
+ return coeffs,modes, umean
+#
+#
+# ------------------------------ Start Class Definition -----------------
+#
+#
class Elbow_profile():
def loadCoordinates(self,case,delete_last=False):
@@ -201,15 +199,10 @@ class Elbow_profile():
# Nrk = self.Rk.size
# Nt = self.dist.size * self.Rk.size
#
- pod = loadPODres(case)
- self.Nmodes = pod["modes"].shape[-1]
- # die moden aufteilen oder nicht??
- self.modes = pod["modes"]
- # the mean profiles
- self.ux_mean = pod_x["umean"]
- self.uy_mean = pod_y["umean"]
- self.uz_mean = pod_z["umean"]
-
+ coeffs, self.modes, self.umean = loadPODres(case)
+ self.Nmodes = self.modes.shape[-1]
+ # split the modes in x, y, z ? better not because of the reconstruction?
+ #
if case=="SingleElbow":
# save the modes in local variables and add zeros at the wall coordinates
# self.modes_ux = np.concatenate((pod_x["modes"].reshape(self.Nphi,self.Nr-1,self.Nmodes),np.zeros((self.Nphi,1,self.Nmodes))),axis=1).reshape(-1,self.Nmodes)
@@ -220,10 +213,7 @@ class Elbow_profile():
# self.uy_mean = np.concatenate((pod_y["umean"].reshape(self.Nphi,self.Nr-1),np.zeros((self.Nphi,1))),axis=1).ravel()
# self.uz_mean = np.concatenate((pod_z["umean"].reshape(self.Nphi,self.Nr-1),np.zeros((self.Nphi,1))),axis=1).ravel()
#
- #
- coeffs = pod_x["coeffs"].reshape((len(self.Rk),len(self.dist),self.Nmodes))
- coeffs_y = pod_y["coeffs"].reshape((len(self.Rk),len(self.dist),self.Nmodes))
- coeffs_z = pod_z["coeffs"].reshape((len(self.Rk),len(self.dist),self.Nmodes))
+ coeffs = coeffs.reshape((len(self.Rk),len(self.dist),self.Nmodes))
else:
# self.modes_ux = pod_x["modes"]
# self.modes_uy = pod_y["modes"]
@@ -233,9 +223,7 @@ class Elbow_profile():
# self.uy_mean = pod_y["umean"]
# self.uz_mean = pod_z["umean"]
#
- coeffs_x = pod_x["coeffs"].reshape((len(self.Rk),len(self.dl),len(self.dist),self.Nmodes))
- coeffs_y = pod_y["coeffs"].reshape((len(self.Rk),len(self.dl),len(self.dist),self.Nmodes))
- coeffs_z = pod_z["coeffs"].reshape((len(self.Rk),len(self.dl),len(self.dist),self.Nmodes))
+ coeffs = coeffs.reshape((len(self.Rk),len(self.dl),len(self.dist),self.Nmodes))
#
# plot for test
# indrk = 7
@@ -249,43 +237,23 @@ class Elbow_profile():
# nicecontour(self.x,self.y,utest)
#polydeg = 3
- self.regint_ux = []
- self.regint_uy = []
- self.regint_uz = []
+ # self.regint_ux = []
+ # self.regint_uy = []
+ # self.regint_uz = []
methodstr = 'linear' # 'linear' 'cubic' or 'quintic' are possible
# the spline interpolation attention it can only evaluate ordered input arrays
# RegularGridInterpolator can also handle multiple output values --> all modes a loop is not neccessary!!!!!
- # to do: Remove the loops, regint is not a list but a interpolater object
+ # but only for linear interpolation not for higher order polynomials!
if case=="SingleElbow":
- for i in range(0,self.Nmodes):
- # self.splineint_ux.append(interpol.RectBivariateSpline(self.dist,self.Rk,self.dl,pod_x["coeffs"][:,:,i].T,kx= polydeg,ky=polydeg ))
- # self.splineint_uy.append(interpol.RectBivariateSpline(self.dist,self.Rk,self.dl,pod_y["coeffs"][:,:,i].T,kx= polydeg,ky=polydeg ))
- # self.splineint_uz.append(interpol.RectBivariateSpline(self.dist,self.Rk,self.dl,pod_z["coeffs"][:,:,i].T,kx= polydeg,ky=polydeg ))
- self.regint_ux.append(interpol.RegularGridInterpolator((self.Rk,self.dist), coeffs_x[:,:,i],method = methodstr, bounds_error = False,fill_value=None))#,bounds_error = False,fill_value=None ))
- self.regint_uy.append(interpol.RegularGridInterpolator((self.Rk,self.dist), coeffs_y[:,:,i],method = methodstr, bounds_error = False,fill_value=None))#,bounds_error = True,fill_value=None ))
- self.regint_uz.append(interpol.RegularGridInterpolator((self.Rk,self.dist), coeffs_z[:,:,i],method = methodstr, bounds_error = False,fill_value=None))#,bounds_error = True,fill_value=None ))
+ self.regint_A = interpol.RegularGridInterpolator((self.Rk,self.dist), coeffs,method = methodstr, bounds_error = False,fill_value=None)
else:
- for i in range(0,self.Nmodes):
- # self.splineint_ux.append(interpol.RectBivariateSpline(self.dist,self.Rk,self.dl,pod_x["coeffs"][:,:,i].T,kx= polydeg,ky=polydeg ))
- # self.splineint_uy.append(interpol.RectBivariateSpline(self.dist,self.Rk,self.dl,pod_y["coeffs"][:,:,i].T,kx= polydeg,ky=polydeg ))
- # self.splineint_uz.append(interpol.RectBivariateSpline(self.dist,self.Rk,self.dl,pod_z["coeffs"][:,:,i].T,kx= polydeg,ky=polydeg ))
- self.regint_ux.append(interpol.RegularGridInterpolator((self.Rk,self.dl,self.dist), coeffs_x[:,:,:,i],method = methodstr ,bounds_error = False,fill_value=None ))
- self.regint_uy.append(interpol.RegularGridInterpolator((self.Rk,self.dl,self.dist), coeffs_y[:,:,:,i],method = methodstr ,bounds_error = False,fill_value=None ))
- self.regint_uz.append(interpol.RegularGridInterpolator((self.Rk,self.dl,self.dist), coeffs_z[:,:,:,i],method = methodstr ,bounds_error = False,fill_value=None ))
- #self.coeffs_z = coeffs_z
- del coeffs_x
- del coeffs_y
- del coeffs_z
+ self.regint_A = interpol.RegularGridInterpolator((self.Rk,self.dl,self.dist), coeffs,method = methodstr ,bounds_error = False,fill_value=None )
#
+ del coeffs
#
- #self.pod_z = pod_z
- del pod_x
- del pod_y
- del pod_z
-
#
self.int_weights = int2d.get_int_weights(x=self.x,y=self.y,method="tri")
-
+ #
self.make_fully()
# The parametrized function to be plotted
#self.get_profile = interpol.RegularGridInterpolator((self.Rk,self.dist), self.uall + self.u_fully)
@@ -297,38 +265,27 @@ class Elbow_profile():
# to evaluate the spline interpolation and get the coefficient matrix at the desired
# values of dist and Rk
if self.case=="SingleElbow":
- A_ux = np.array([int_i((Rk,dist)).T for int_i in self.regint_ux] ).T
- A_uy = np.array([int_i((Rk,dist)).T for int_i in self.regint_uy] ).T
- A_uz = np.array([int_i((Rk,dist)).T for int_i in self.regint_uz] ).T
+ A = self.regint_A((Rk,dist))
+ # A_uz = np.array([int_i((Rk,dist)).T for int_i in self.regint_uz] ).T
else:
- A_ux = np.array([int_i((Rk,dl,dist)).T for int_i in self.regint_ux] ).T
- A_uy = np.array([int_i((Rk,dl,dist)).T for int_i in self.regint_uy] ).T
- A_uz = np.array([int_i((Rk,dl,dist)).T for int_i in self.regint_uz] ).T
+ A = self.regint_A((Rk,dl,dist))
+ # A_ux = np.array([int_i((Rk,dl,dist)).T for int_i in self.regint_ux] ).T
#
# print("this is A for dist = " + str(dist) + " and Rk = " + str(Rk) + ":")
# print(A_uz)
#
- A_ux = A_ux.reshape(-1,A_ux.shape[-1])
- A_uy = A_uy.reshape(-1,A_uy.shape[-1])
- A_uz = A_uz.reshape(-1,A_uz.shape[-1])
- # now reconstruct the velocities
- # print(A_ux.shape)
+ A = A.reshape(-1,A.shape[-1])
# print(A_uy.shape)
# print(A_uz.shape)
if asmat:
- ux = do_reconstruct(A_ux, self.modes_ux, umean = self.ux_mean).reshape(self.xm.shape)
- uy = do_reconstruct(A_uy, self.modes_uy, umean = self.uy_mean).reshape(self.xm.shape)
- uz = do_reconstruct(A_uz, self.modes_uz, umean = self.uz_mean).reshape(self.xm.shape)
- # if the dimensions of a is larger:
- # b1 = bla.reshape(*bla.shape[0:-1],self.xm.shape)
+ # reshape to u.shape = 3 * x.shape
+ u = do_reconstruct(A,self.modes,umean = self.umean ).reshape(3,*self.xm.shape)
else:
- ux = do_reconstruct(A_ux, self.modes_ux, umean = self.ux_mean)
- uy = do_reconstruct(A_uy, self.modes_uy, umean = self.uy_mean)
- uz = do_reconstruct(A_uz, self.modes_uz, umean = self.uz_mean)
+ u = do_reconstruct(A, self.modes, umean = self.umean)
if addfully:
- uz += self.u_fully
+ u[2] += self.u_fully
#
- return ux,uy,uz
+ return u
def get_path_profile_diam_se(self,Rk,dist,reverse =False, addfully = False,R = 1):
# returns the velocity profile at a certain distance and coordinates x,y, (if given)
--
GitLab