from emcee import PTSampler import mkl mkl.set_num_threads(1) import time from pztel_constants import constants from datetime import datetime,date #import jdcal import pdb #from orbitclass import orbit import matplotlib.pyplot as plt import pickle as pickle from scipy.integrate import odeint from RK4_energy import * G = 4.*np.pi**2. parallax,distance,mstar,kgauss, AU, DAY = constants() savename = '/big_scr7/dryan/pztelmc/pztel_cart_reproc1' #savename = 'cart_test2' filename = 'data_pztel_reproc.csv' eps=(np.genfromtxt(filename, skip_header=1,delimiter=',', usecols=0, unpack=True)).T.ravel() x=(np.genfromtxt(filename, skip_header=1,delimiter=',', usecols=7, unpack=True)).T.ravel()/1000 y=(np.genfromtxt(filename, skip_header=1,delimiter=',', usecols=8, unpack=True)).T.ravel()/1000 sigx=(np.genfromtxt(filename, skip_header=1,delimiter=',', usecols=9, unpack=True)).T.ravel()/1000 sigy=(np.genfromtxt(filename, skip_header=1,delimiter=',', usecols=10, unpack=True)).T.ravel()/1000 eps=eps[x == x]/365.2425 sigx=sigx[x==x] sigy=sigy[x==x] y=y[x==x] x=x[x==x]
def koe(epochs, a, tau, argp, lan, inc, ecc, mass, para):
# Epochs in MJD
#
# date twice but x & y are computed.
# The data are returned so that the values in the array
# alternate x and y pairs.
#
# Stellar properties for beta Pic
parallax, distance, mstar, kgauss, AU, DAY = constants()
mstar = mass
parallax = para
distance = 1. / parallax
ndate = len(epochs)
# Keplerian Elements
#
# epochs dates in modified JD [day]
# a semimajor axis [au]
# tau epoch of peri in units of the orbital period
# argp argument of peri [radians]
# lan longitude of ascending node [radians]
# inc inclination [radians]
# ecc eccentricity
#
# Derived quantities
# manom --- mean anomaly
# eccanom --- eccentric anomaly
# truan --- true anomaly
# theta --- longitude
# radius --- star-planet separation
n = kgauss * np.sqrt(mstar) * (a)**(-1.5) # compute mean motion in rad/day
# ---------------------------------------
# Compute the anomalies (all in radians)
#
# manom = n * (epochs - tau) # mean anomaly
manom = n * epochs[
0::2] - 2 * np.pi * tau # Mean anomaly w/tau in units of period
eccanom = np.array([])
for man in manom:
#print man #DEBUGGING
eccanom = np.append(eccanom, ema(man % (2 * np.pi), ecc))
# ---------------------------------------
# compute the true anomaly and the radius
#
# Elliptical orbit only
truan = 2. * np.arctan(
np.sqrt((1.0 + ecc) / (1.0 - ecc)) * np.tan(0.5 * eccanom))
theta = truan + argp
radius = a * (1.0 - ecc * np.cos(eccanom))
# ---------------------------------------
# Compute the vector components in
# ecliptic cartesian coordinates (normal convention Green Eq. 7.7)
# standard form
#
#xp = radius *(np.cos(theta)*np.cos(lan) - np.sin(theta)*np.sin(lan)*np.cos(inc))
#yp = radius *(np.cos(theta)*np.sin(lan) + np.sin(theta)*np.cos(lan)*np.cos(inc))
#
# write in terms of \Omega +/- omega --- see Mathematica notebook trig-id.nb
c2i2 = np.cos(0.5 * inc)**2
s2i2 = np.sin(0.5 * inc)**2
arg0 = truan + lan
arg1 = truan + argp + lan
arg2 = truan + argp - lan
arg3 = truan - lan
c1 = np.cos(arg1)
c2 = np.cos(arg2)
s1 = np.sin(arg1)
s2 = np.sin(arg2)
sa0 = np.sin(arg0)
sa3 = np.sin(arg3)
# updated sign convention for Green Eq. 19.4-19.7
xp = radius * (c2i2 * s1 - s2i2 * s2)
yp = radius * (c2i2 * c1 + s2i2 * c2)
# Interleave x & y
# put x data in odd elements and y data in even elements
data = np.zeros(ndate)
data[0::2] = xp
data[1::2] = yp
return data * parallax # results in seconds of arc