From 6118512bee44db001d4dd0a3046bad2aecf0da30 Mon Sep 17 00:00:00 2001 From: johannesring Date: Mon, 19 Jun 2023 12:41:44 +0200 Subject: [PATCH] np.complex -> np.complex128 (fixes issue #69) --- turtleFSI/utils/Womersley.py | 10 +++++----- 1 file changed, 5 insertions(+), 5 deletions(-) diff --git a/turtleFSI/utils/Womersley.py b/turtleFSI/utils/Womersley.py index 2f3bce4..80e68f1 100755 --- a/turtleFSI/utils/Womersley.py +++ b/turtleFSI/utils/Womersley.py @@ -167,10 +167,10 @@ def _precompute_bessel_functions(self): self.ns = np.arange(1, self.N) # Allocate for 0...N-1 - alpha = np.zeros(self.N, dtype=np.complex) - self.beta = np.zeros(self.N, dtype=np.complex) - self.jn0_betas = np.zeros(self.N, dtype=np.complex) - self.jn1_betas = np.zeros(self.N, dtype=np.complex) + alpha = np.zeros(self.N, dtype=np.complex128) + self.beta = np.zeros(self.N, dtype=np.complex128) + self.jn0_betas = np.zeros(self.N, dtype=np.complex128) + self.jn1_betas = np.zeros(self.N, dtype=np.complex128) # Compute vectorized for 1...N-1 (keeping element 0 in arrays to make indexing work out later) alpha[1:] = self.radius * np.sqrt(self.ns * (self.omega / self.nu)) @@ -181,7 +181,7 @@ def _precompute_bessel_functions(self): def _precompute_r_dependent_coeffs(self, y): pir2 = np.pi * self.radius**2 # Compute intermediate terms for womersley function - r_dependent_coeffs = np.zeros(self.N, dtype=np.complex) + r_dependent_coeffs = np.zeros(self.N, dtype=np.complex128) if hasattr(self, 'Vn'): #r_dependent_coeffs[0] = (self.Vn[0]/2.0) * (1 - y**2) r_dependent_coeffs[0] = self.Vn[0] * (1 - y**2)