artiq/examples/master/repository/flopping_f_simulation.py

68 lines
2.2 KiB
Python
Raw Normal View History

2015-01-13 19:12:35 +08:00
from math import sqrt, cos, pi
import time
import random
import numpy as np
from scipy.optimize import curve_fit
2015-01-13 19:12:35 +08:00
from artiq import *
def model(x, F0):
t = 0.02
tpi = 0.03
A = 80
B = 40
2015-01-13 19:12:35 +08:00
return A+(B-A)/2/(4*tpi**2*(x-F0)**2+1)*(1-cos(pi*t/tpi*sqrt(4*tpi**2*(x-F0)**2+1)))
def model_numpy(xdata, F0):
r = np.zeros(len(xdata))
for i, x in enumerate(xdata):
r[i] = model(x, F0)
return r
2015-07-14 04:08:20 +08:00
class FloppingF(EnvExperiment):
"""Flopping F simulation"""
2015-07-14 04:08:20 +08:00
def build(self):
self.attr_argument("frequency_scan", Scannable(
default=LinearScan(1000, 2000, 100)))
2015-01-13 19:12:35 +08:00
self.attr_argument("F0", NumberValue(1500, min=1000, max=2000))
self.attr_argument("noise_amplitude", NumberValue(0.1, min=0, max=100,
step=0.01))
2015-01-13 19:12:35 +08:00
2015-07-14 04:08:20 +08:00
self.attr_device("scheduler")
2015-01-13 19:12:35 +08:00
def run(self):
frequency = self.set_result("flopping_f_frequency", [],
realtime=True, store=False)
brightness = self.set_result("flopping_f_brightness", [],
realtime=True)
self.set_result("flopping_f_fit", [], realtime=True, store=False)
for f in self.frequency_scan:
m_brightness = model(f, self.F0) + self.noise_amplitude*random.random()
frequency.append(f)
brightness.append(m_brightness)
2015-01-13 19:12:35 +08:00
time.sleep(0.1)
2015-05-17 16:11:00 +08:00
self.scheduler.submit(self.scheduler.pipeline_name, self.scheduler.expid,
2015-05-28 17:20:58 +08:00
self.scheduler.priority, time.time() + 20, False)
2015-01-13 19:12:35 +08:00
def analyze(self):
# Use get_result so that analyze can be run stand-alone.
frequency = self.get_result("flopping_f_frequency")
brightness = self.get_result("flopping_f_brightness")
popt, pcov = curve_fit(model_numpy,
frequency, brightness,
2015-07-14 04:08:20 +08:00
p0=[self.get_parameter("flopping_freq")])
perr = np.sqrt(np.diag(pcov))
if perr < 0.1:
F0 = float(popt)
self.set_parameter("flopping_freq", F0)
self.set_result("flopping_f_fit",
[model(x, F0) for x in frequency],
realtime=True, store=False)