def effectiv_plot(self, data, I_rms, U_rms, R, P, data_name: str,
save_pic: bool, plot_pic: bool, zoom: bool,
save_pic_path: str):
'''
----------------
DESCRIPTION
----------------
A function to plot the effectiv data:
I_rms: effectiv current
U_rms: effectiv voltage
R: resistance
P: power
----------------
PARAMETER
----------------
data_name - name of the current data
peak_id - peak index of the current
valley_id - valley index of the current
I_rms - effectiv current
U_rms - effectiv voltage
R - resistance
p - power
plot_pic - bool
zoom - bool show local zoom of current data
save_pic - bool save the pic or not
save_pic_path - if save pic is True, please gave the save path
'''
begin_id, end_id = self.begin_id, self.end_id
valley_id, peak_id = self.valley_id, self.peak_id
font = {
'family': 'serif',
# 'color': 'darkred',
'weight': 'normal',
'size': 12
}
prop = {'family': 'serif', 'size': 10}
with plt.rc_context({'axes.autolimit_mode': 'round_numbers'}):
fig = plt.figure(constrained_layout=False, figsize=(10, 6))
gs = fig.add_gridspec(nrows=2, ncols=2, wspace=0.2)
t = data.Zeit[begin_id:end_id]
####current
ax1 = fig.add_subplot(gs[0, 0])
ax1.plot(t,
data.Schweissstrom[begin_id:end_id],
color='silver',
label='Schweißstrom roh')
ax1.plot(t,
I_rms[begin_id:end_id],
color='k',
label='effektiver Schweißstrom')
# ax1.scatter(data.loc[:,'Zeit'][peak_id], data.loc[:,'Schweissstrom'][peak_id], color = 'red', edgecolors = 'k', marker = 'o', alpha = 0.5, label = 'Peak')
# ax1.scatter(data.loc[:,'Zeit'][valley_id], data.loc[:,'Schweissstrom'][valley_id], color = 'green', edgecolors = 'k', marker = 's', alpha = 0.5, label = 'Tal')
ax1.legend(loc='best', prop=prop)
# ax1.legend(loc='upper center', bbox_to_anchor=(0.5,1.2),frameon=False,ncol=2,prop = prop)
ax1.set_ylabel('Schweissstrom $I_{s}$ / $kA$', fontdict=font)
ax1.set_xlim(1.1, 1.9)
# ax1.tick_params(axis='both',labelsize = 8,direction = 'in')
# ax1.set_title('Schweißstrom',fontdict = font)
# ax1.set_xlabel('Zeit $t$ / $s$',fontdict = font)
ax1.tick_params(axis='both',
which='major',
labelsize=12,
direction='in')
ax1.grid()
plt.setp(ax1.get_xticklabels(), visible=False)
if zoom:
with plt.rc_context({'axes.autolimit_mode': 'round_numbers'}):
axins = ax1.inset_axes((0.2, 0.2, 0.45, 0.3))
axins.plot(t,
data.Schweissstrom[begin_id:end_id],
color='silver')
axins.plot(t, I_rms[begin_id:end_id], color='k')
axins.scatter(data.loc[:, 'Zeit'][peak_id],
data.loc[:, 'Schweissstrom'][peak_id],
color='red',
edgecolors='k',
marker='o',
alpha=0.5)
axins.scatter(data.loc[:, 'Zeit'][valley_id],
data.loc[:, 'Schweissstrom'][valley_id],
color='green',
edgecolors='k',
marker='s',
alpha=0.5)
zone_left = begin_id + 8000
zone_right = zone_left + 500
x_ratio = 0
y_ratio = 0.05
x_ = data.loc[:, 'Zeit']
xlim0 = x_[zone_left] - (x_[zone_right] -
x_[zone_left]) * x_ratio
xlim1 = x_[zone_right] + (x_[zone_right] -
x_[zone_left]) * x_ratio
y = data.loc[zone_left:zone_right, 'Schweissstrom']
ylim0 = np.min(y) - (np.max(y) - np.min(y)) * y_ratio - 0.5
ylim1 = np.max(y) + (np.max(y) - np.min(y)) * y_ratio + 0.5
axins.set_xlim(xlim0, xlim1)
axins.set_ylim(ylim0, ylim1)
axins.set_xticks([])
axins.set_yticks([])
tx0 = xlim0
tx1 = xlim1 + 0.01
ty0 = ylim0
ty1 = ylim1
sx = [tx0, tx1, tx1, tx0, tx0]
sy = [ty0, ty0, ty1, ty1, ty0]
ax1.plot(sx, sy, "darkred", alpha=0.5)
xy = (xlim0, ylim0)
xy2 = (xlim0, ylim1)
con = ConnectionPatch(xyA=xy2,
xyB=xy,
coordsA="data",
coordsB="data",
axesA=axins,
axesB=ax1,
color='darkred',
alpha=0.5)
axins.add_artist(con)
xy = (xlim1, ylim0)
xy2 = (xlim1, ylim1)
con = ConnectionPatch(xyA=xy2,
xyB=xy,
coordsA="data",
coordsB="data",
axesA=axins,
axesB=ax1,
color='darkred',
alpha=0.5)
axins.add_artist(con)
# voltage
ax2 = fig.add_subplot(gs[1, 0], sharex=ax1)
ax2.plot(t,
data.Schweissspannung[begin_id:end_id],
color='silver',
label='Schweißspannung roh')
ax2.plot(t,
U_rms[begin_id:end_id],
color='k',
label='effektive Schweißspannung')
# ax2.scatter(data.loc[:,'Zeit'][valley_id], data.loc[:,'Schweissspannung'][valley_id], color = 'green', edgecolors = 'k', marker = 's', alpha = 0.5, label = 'Tal')
ax2.legend(loc='best', prop=prop)
# ax2.legend(loc='upper center', bbox_to_anchor=(0.5,1.2),frameon=False,ncol=2,prop = prop)
ax2.set_ylabel('Schweißspannung $U_{s}$ / $V$', fontdict=font)
ax2.set_xlabel('Zeit $t$ / $s$', fontdict=font)
# ax2.tick_params(axis='both',labelsize = 8,direction = 'in')
ax2.tick_params(axis='both',
which='major',
labelsize=12,
direction='in')
# ax2.set_title('Schweißspannung',fontdict = font)
ax2.grid()
plt.setp(ax2.get_xticklabels(), visible=True)
if zoom:
with plt.rc_context({'axes.autolimit_mode': 'round_numbers'}):
axins2 = ax2.inset_axes((0.3, 0.2, 0.45, 0.3))
axins2.plot(t,
data.Schweissspannung[begin_id:end_id],
color='silver')
axins2.plot(t, U_rms[begin_id:end_id], color='k')
# axins2.scatter(data.loc[:,'Zeit'][peak_id], data.loc[:,'Schweissspannung'][peak_id], color = 'red', edgecolors = 'k', marker = 'o', alpha = 0.3)
axins2.scatter(data.loc[:, 'Zeit'][valley_id],
data.loc[:, 'Schweissspannung'][valley_id],
color='green',
edgecolors='k',
marker='s',
alpha=0.5)
zone_left = begin_id + 8000
zone_right = zone_left + 500
x_ratio = 0
y_ratio = 0.05
x_ = data.loc[:, 'Zeit']
xlim0 = x_[zone_left] - (x_[zone_right] -
x_[zone_left]) * x_ratio
xlim1 = x_[zone_right] + (x_[zone_right] -
x_[zone_left]) * x_ratio
y = data.loc[zone_left:zone_right, 'Schweissspannung']
ylim0 = np.min(y) - (np.max(y) - np.min(y)) * y_ratio - 0.5
ylim1 = np.max(y) + (np.max(y) - np.min(y)) * y_ratio + 0.5
axins2.set_xlim(xlim0, xlim1)
axins2.set_ylim(ylim0, ylim1)
axins2.set_xticks([])
axins2.set_yticks([])
tx0 = xlim0
tx1 = xlim1 + 0.01
ty0 = ylim0
ty1 = ylim1
sx = [tx0, tx1, tx1, tx0, tx0]
sy = [ty0, ty0, ty1, ty1, ty0]
ax2.plot(sx, sy, "darkred", alpha=0.5)
xy = (xlim0, ylim0)
xy2 = (xlim0, ylim0)
con = ConnectionPatch(xyA=xy2,
xyB=xy,
coordsA="data",
coordsB="data",
axesA=axins2,
axesB=ax2,
color='darkred',
alpha=0.5)
axins2.add_artist(con)
xy = (xlim1, ylim0)
xy2 = (xlim1, ylim0)
con = ConnectionPatch(xyA=xy2,
xyB=xy,
coordsA="data",
coordsB="data",
axesA=axins2,
axesB=ax2,
color='darkred',
alpha=0.5)
axins2.add_artist(con)
# resistance
ax3 = fig.add_subplot(gs[0, 1])
ax3.plot(t,
R[begin_id:end_id],
color='k',
label='dynamischer Widerstand')
# ax3.legend(loc = 1,prop = prop,frameon=False)
ax3.set_ylabel('dynamischer Widerstand $R$ / $mΩ$', fontdict=font)
# ax3.text(max(t)*0.7,max(R[begin_id:end_id]),'$R = U_{eff} / I_{eff}$',fontdict=font)
ax3.tick_params(axis='both', labelsize=12, direction='in')
ax3.set_xlim(1.1, 1.9)
# ax3.set_title('Prozesswiderstand',fontdict = font)
# ax3.set_xlabel('Zeit $t$ / $s$',fontdict = font)
ax3.grid()
plt.setp(ax3.get_xticklabels(), visible=False)
# power
ax4 = fig.add_subplot(gs[1, 1], sharex=ax3)
ax4.plot(t, P[begin_id:end_id], color='k', label='Prozessleistung')
# ax4.legend(loc = 'best',prop = prop,frameon=False)
ax4.set_ylabel('Prozessleistung $P$/ $kW$', fontdict=font)
ax4.set_xlabel('Zeit $t$ / $s$', fontdict=font)
# ax4.set_title('Prozessleistung',fontdict = font)
# ax4.text(max(t)*0.7,max(P[begin_id:end_id]),'$P = I^{2}_{eff} · R$',fontdict=font)
ax4.tick_params(axis='both', labelsize=12, direction='in')
ax4.grid()
plt.setp(ax4.get_xticklabels(), visible=True)
# fig.suptitle('{}'.format(data_name), fontdict = font,y = 0.95, fontsize = 10)
if save_pic:
save_fig(image_path=save_pic_path, fig_name=data_name)
print(save_pic_path)
if plot_pic:
plt.show()
def change_point(self,
width: int,
cut_off: list,
custom_cost,
jump: int,
pen: float,
results_show: bool,
title=None,
save_path=None,
fig_name=None):
'''
----------------
DESCRIPTION
----------------
The purpose of the change point detection is to check whether there is a large enough sudden change
in a specific interval interval of the resistance signal.
If there is a large enough change, it means that the explosion phenomenon has occurred during this welding.
the algorithms of detection can be fund by this link:
https://centre-borelli.github.io/ruptures-docs/index.html#documentation
for the resistance data especially MDK2 this methode can be used to detective if a change point in selectarea
if there is a change point, mean value before change point and after change point will be compared --> delta R
else no change point delta R = 0
because of material loss the dalta R musst bigger than 0, if the there is a chagne point but delta R < 0,
this situation has nothing to do with spritzer rarely occurs delta R can also be 0
and as usaual the resistance curve is going down with the time
----------------
PARAMETER
----------------
width: int windows width 40
cutoff: list [float, float], float: 0...1 1 means all data length will be selected [0.15, 0.45]
custom_cost: https://centre-borelli.github.io/ruptures-docs/costs/index.html
jump: int subsample (one every jump points) 5
pen: float penalty value (>0) 2
result_show : show image evaluation to displan the detective result
title: the image title
save_path: the path to save the result image
fig_name: the image name
----------------
RETURN
----------------
delta_R: the Variation before and after the change point of resistance signal
'''
ab_R = self.R_data[round(len(self.valley_id) * cut_off[0]
):round(len(self.valley_id) *
cut_off[1])].values
c = custom_cost
algo = rpt.Window(width=width, custom_cost=c,
jump=jump).fit_predict(ab_R, pen=2)
if len(algo) >= 2:
delta_R = np.mean(ab_R[:algo[0]]) - np.mean(ab_R[algo[0]:])
if delta_R < 0:
# delta_R can not less than 0 bescause the the resistance curve is going down with the time
delta_R = 0
else:
delta_R = 0
if results_show:
rpt.display(ab_R, algo)
if title != None:
plt.title(title)
if save_path and fig_name is not None:
save_fig(image_path=save_path, fig_name=fig_name)
plt.show()
return delta_R
ax3.set_xlabel('Festerbreite',fontdict = font)
ax3.set_title('Bias$^2$',fontdict = font)
ax4 = fig.add_subplot(232)
ax4.plot(EF_arr, Var,color = 'gray',linestyle = '--')
ax4.scatter(EF_arr, Var, c=Var, cmap = cm.viridis)
ax4.set_xlim(1,max(EF_arr))
ax4.yaxis.get_major_formatter().set_powerlimits((0,1))
# ax3.set_ylim(min(Var)*0.99,max(Var)*1.01)
ax4.tick_params(axis='both', which='major', labelsize=12,direction='in')
ax4.set_xlabel('Festerbreite',fontdict = font)
ax4.set_title('Varianz',fontdict = font)
ax5 = fig.add_subplot(233)
ax5.plot(EF_arr, min_ab_l,color = 'gray',linestyle = '--')
ax5.scatter(EF_arr, min_ab_l, c=min_ab_l, cmap = cm.viridis)
ax5.set_xlim(1,max(EF_arr))
ax5.yaxis.get_major_formatter().set_powerlimits((0,1))
# ax4.set_ylim(min(min_ab_l)*0.99,max(min_ab_l)*1.01)
ax5.tick_params(axis='both', which='major', labelsize=12,direction='in')
ax5.set_xlabel('Festerbreite',fontdict = font )
ax5.set_title('Abweichung',fontdict = font)
fig.canvas.manager.window.move(0,0)
plt.subplots_adjust(left=None, bottom=None, right=None, top=None,
wspace=None, hspace=0.3)
save_fig(image_path = r_path,fig_name = 'Beste_Fensterbreite_einzel_2')
plt.show()
index.remove(target_col)
index_ranking = abs(
corr_matrix.loc[index, target_col]).sort_values(ascending=False)
val_ranking = sorted(corr_matrix.loc[index, target_col],
key=abs,
reverse=True)
corr_ranking = pd.DataFrame(val_ranking,
columns=[target_col],
index=index_ranking.index)
return corr_ranking
corr_ranking(all_data, 'MEDV').plot.bar()
plt.title('Korrelationskoeffizient zum Target')
plt.grid(axis='y')
figsave.save_fig(IMAGES_PATH, 'Korrelationskoeffizient zum Target')
plt.show()
# # split the data to traindata and test data
X_train, X_test, y_train, y_test = train_test_split(X_all_data,
y_all_targe,
test_size=test_size,
random_state=my_seed)
# data normalisation
scaler = StandardScaler()
X_train_nor = pd.DataFrame(scaler.fit_transform(X_train.values),
index=X_train.index,
columns=X_train.columns)
X_test_nor = pd.DataFrame(scaler.transform(X_test.values),
index=X_test.index,
columns=X_test.columns)
ax6.set_yticks([])
ax6.set_title('Q = {}'.format(Q_l[11]), fontdict=font)
left, bottom, width, height = 0.3, 0.7, 0.2, 0.15
ax7 = fig.add_axes([left, bottom, width, height])
ax7.plot(t, w_U, label='OG', color='gray')
ax7.plot(t, filenames_list_all[Q_l[13]][0], label='filted', color='k')
ax7.set_xticks([])
ax7.set_yticks([])
ax7.set_title('Q = {}'.format(Q_l[13]), fontdict=font)
left, bottom, width, height = 0.5, 0.7, 0.2, 0.15
ax8 = fig.add_axes([left, bottom, width, height])
ax8.plot(t, w_U, label='OG', color='gray')
ax8.plot(t, filenames_list_all[Q_l[17]][0], label='filted', color='k')
ax8.set_xticks([])
ax8.set_yticks([])
ax8.set_title('Q = {}'.format(Q_l[17]), fontdict=font)
left, bottom, width, height = 0.7, 0.7, 0.2, 0.15
ax9 = fig.add_axes([left, bottom, width, height])
ax9.plot(t, w_U, label='OG', color='gray')
ax9.plot(t, filenames_list_all[Q_l[19]][0], label='Kalman', color='k')
ax9.set_xticks([])
ax9.set_yticks([])
# ax9.legend(prop = legend_properties,loc = 4 )
ax9.set_title('Q = {}'.format(Q_l[19]), fontdict=font)
save_fig(image_path=s_path, fig_name='Kalmanfilter_Q_3')
plt.show()
fig, ax = plt.subplots(figsize=(5, 4))
width = 0.6
index = y.index
y_ = y.reset_index(drop=True)
ax.bar(y_.index, y_.values, width, color='gray')
ax.bar(y_.index[-1], y_.values[-1], width, color='darkred')
ax.set_ylim(0.7, 0.8)
ax.set_xticks(np.arange(0, 32, 5))
ax.set_ylabel('Korrelationskoeffizient', fontdict=font)
ax.set_xlabel('partielle Wärmemenge Index', fontdict=font)
ax.set_xticklabels(index[::5], fontdict=font)
ax.tick_params(axis='y', which='major', labelsize=12, direction='in')
ax.tick_params(axis='x', rotation=45, which='major', labelsize=12)
ax.scatter(0, 0, alpha=0, label='$Q_{i}$ - ges. Wärmemenge je 20 ms')
plt.legend(frameon=False, prop=prop)
save_fig(image_path='C:/DA/Code/pywt/images/MDK2/',
fig_name='dQ_dw_2',
reselution=200)
plt.show()
# fig, ax = plt.subplots(figsize=(15,10))
# ax = sn.heatmap(data_transposed.corr(),ax = ax,annot = True,linewidths=.5,cbar_kws= {'label': 'Korrlationskoeffizient','aspect':40,'pad':0.01})
# ax.figure.axes[-1].yaxis.label.set_size(14)
# ax.tick_params(labelsize = 10)
# ax.set_ylabel('Parameternummer', fontsize = 14)
# ax.set_xlabel('Parameternummer', fontsize = 14)
# # save_fig(image_path=s_path, fig_name='Korrelation_Q_dw_2')
# plt.show()
# print(data_transposed.corr())