From b18119106ff8023c06edbc7970eb3659ec2bbf59 Mon Sep 17 00:00:00 2001
From: Martin Killenberg <martin.killenberg@desy.de>
Date: Thu, 26 Oct 2023 16:11:06 +0200
Subject: [PATCH] feat: post analysis for RF cables
---
Python_script/analysis.py | 70 +++++++++++++++++++++++++++++++++------
1 file changed, 59 insertions(+), 11 deletions(-)
diff --git a/Python_script/analysis.py b/Python_script/analysis.py
index 2fa4aaa..94bbf77 100644
--- a/Python_script/analysis.py
+++ b/Python_script/analysis.py
@@ -3,7 +3,7 @@ import matplotlib.pyplot as plt
import numpy as np
from matplotlib import gridspec
-def extract_stable_data(datafile, measurement_set, reference_signal_names):
+def extract_stable_data(datafile, measurement_set, reference_signal_names, extra_signal_names):
datapoint = {}
# df is a pandas data frame
df = pd.read_csv(datafile)
@@ -29,23 +29,32 @@ def extract_stable_data(datafile, measurement_set, reference_signal_names):
datapoint['humidity_mean'] = humidities.mean()
datapoint['humidity_var'] = humidities.var()
+ for extra_signal_name in extra_signal_names:
+ extra_signal_values = df.loc[(df['EQUILIBRIUM_INDICATOR'] == 31) & (df['SET_NAME'] == measurement_set), extra_signal_name]
+ datapoint[extra_signal_name] = extra_signal_values.mean()
+
return datapoint
# sweep_type is either 'temperature' or 'humidity'
-def plot_sweep(temperatures, humidities, basename, sweep_type, measurement_sets, reference_signal_names, normalise = [True, False]):
+def plot_sweep(temperatures, humidities, basename, sweep_type, measurement_sets, reference_signal_names,
+ analysis_config = {'type': 'rf_cable', 'normalise': [True, False], 'cable_length': 10,
+ 'extra_signal_names' : ['RF_FREQUENCY']}):
set_data = {}
derivatives = {}
for measurement_set in measurement_sets:
set_data[measurement_set] = {'signal0_means': [], 'signal0_vars': [], 'signal1_means': [], 'signal1_vars': [],
'x_data': []}
+ for external_signal_name in analysis_config['extra_signal_names']:
+ set_data[measurement_set][external_signal_name] = []
derivatives[measurement_set] = {'signal0_deltas': [], 'signal1_deltas': [], 'x_deltas': []}
-
+
for temp, hum in zip(temperatures, humidities):
datafile = basename+'_'+str(temp)+'deg_'+str(hum)+'rh.csv'
print(datafile)
for measurement_set in measurement_sets:
- datapoint = extract_stable_data(datafile, measurement_set, reference_signal_names)
+ datapoint = extract_stable_data(datafile, measurement_set, reference_signal_names,
+ analysis_config['extra_signal_names'])
if datapoint is None:
continue
@@ -62,11 +71,14 @@ def plot_sweep(temperatures, humidities, basename, sweep_type, measurement_sets,
data['signal1_means'].append(datapoint['signal1_mean'])
data['signal1_vars'].append(datapoint['signal1_var'])
+ for external_signal_name in analysis_config['extra_signal_names']:
+ data[external_signal_name].append(datapoint[external_signal_name])
+
for measurement_set in measurement_sets:
data = set_data[measurement_set]
- if normalise[0]:
+ if analysis_config['normalise'][0]:
data['signal0_means'] -= data['signal0_means'][0]
- if normalise[1]:
+ if analysis_config['normalise'][1]:
data['signal1_means'] -= data['signal1_means'][0]
fig = plt.figure(figsize=(10, 15))
@@ -78,9 +90,9 @@ def plot_sweep(temperatures, humidities, basename, sweep_type, measurement_sets,
fig.text(0.1, 0.9, upper_text_block)
ax1 = fig.add_subplot(gs[1, 0])
- ax2 = fig.add_subplot(gs[2, 0],sharex=ax1)
+ ax2 = fig.add_subplot(gs[2, 0], sharex=ax1)
ax4 = fig.add_subplot(gs[4, 0])
- ax5 = fig.add_subplot(gs[5, 0],sharex=ax4)
+ ax5 = fig.add_subplot(gs[5, 0], sharex=ax4)
ax1.tick_params(bottom=True, top=True, direction='in')
ax2.tick_params(bottom=True, top=True, direction='inout')
@@ -132,12 +144,48 @@ def plot_sweep(temperatures, humidities, basename, sweep_type, measurement_sets,
(data['x_data'][i+1]+data['x_data'][i]))
deriv_data['x_deltas'].append((data['x_data'][i+1]+data['x_data'][i])/2)
- ax4.scatter(derivatives[measurement_set]['x_deltas'], derivatives[measurement_set]['signal0_deltas'],marker='+', )
- ax5.scatter(derivatives[measurement_set]['x_deltas'], derivatives[measurement_set]['signal1_deltas'],marker='+')
-
ax4.set_ylabel('$\\Delta$ '+reference_signal_names[0]+' '+denominator_name)
ax5.set_ylabel('$\\Delta$ '+reference_signal_names[1]+' '+denominator_name)
+ #######################################################################################################################
+ # RF cable post analysis: Normalise to cable length and convert phase to time
+ #######################################################################################################################
+
+ if analysis_config['type'] == 'rf_cable':
+ for measurement_set in measurement_sets:
+ data = set_data[measurement_set]
+ deriv_data = derivatives[measurement_set]
+ phase_to_time = 1e15 / (analysis_config['cable_length'] * data['RF_FREQUENCY'][0] * 360.)
+ deriv_data['signal0_deltas'][:] = [x * phase_to_time for x in deriv_data['signal0_deltas']]
+ deriv_data['signal1_deltas'][:] = [x / analysis_config['cable_length'] for x in deriv_data['signal1_deltas']]
+
+ ax4.set_ylabel('$\\Delta$t [fs/m/K]')
+ ax5.set_ylabel('$\\Delta$A [dB/m/K]')
+
+
+ #######################################################################################################################
+ # Closeout
+ #######################################################################################################################
+ #manual scaling if y axis. Auto is broken with scatter.
+ ax4_mins = []
+ ax4_maxs = []
+ ax5_mins = []
+ ax5_maxs = []
+
+ for measurement_set in measurement_sets:
+ ax4.scatter(derivatives[measurement_set]['x_deltas'], derivatives[measurement_set]['signal0_deltas'], marker='+')
+ ax5.scatter(derivatives[measurement_set]['x_deltas'], derivatives[measurement_set]['signal1_deltas'], marker='+')
+ ax4_mins.append(min(derivatives[measurement_set]['signal0_deltas']))
+ ax4_maxs.append(max(derivatives[measurement_set]['signal0_deltas']))
+ ax5_mins.append(min(derivatives[measurement_set]['signal1_deltas']))
+ ax5_maxs.append(max(derivatives[measurement_set]['signal1_deltas']))
+
+ # set auto y range which does not work properly for small values in scatter
+ dy4 = (max(ax4_maxs) - min(ax4_mins))*0.1
+ ax4.set_ylim(min(ax4_mins)-dy4, max(ax4_maxs)+dy4)
+ dy5 = (max(ax5_maxs) - min(ax5_mins))*0.1
+ ax5.set_ylim(min(ax5_mins)-dy5, max(ax5_maxs)+dy5)
+
fig.tight_layout() # otherwise the legend is clipped
fig.savefig(basename+measurement_set+'_analysis.pdf')
--
GitLab