def hypothesisTest(sensor1, sensor2, vari):
TD, WS1, CwS, HwS, Temp1, GTemp, WC, RH, HSI, DP, PWBTemp, SP, BP, Alt, DAlt, NAWBTemp, WBGT, TWL, MD = readdata(
sensor1)
TD, WS2, CwS, HwS, Temp2, GTemp, WC, RH, HSI, DP, PWBTemp, SP, BP, Alt, DAlt, NAWBTemp, WBGT, TWL, MD = readdata(
sensor2)
if vari == 'T':
A = Temp1
B = Temp2
elif vari == 'WS':
A = WS1
B = WS2
t, p = stats.ttest_ind(A, B)
t = abs(t)
if vari == 'T':
print('t(Temperature) =', t)
print('p-value(Temperature) =', p)
elif vari == 'WS':
print('t(Wind speed) =', t)
print('p-value(Wind speed) =', p)
def plotScatter(A,B,C,D,E,vari):
TD, WS, CwSA, HwSA, TempA, GTemp, WC, RH, HSI, DP, PWBTempA, SP, BP, Alt, DAlt, NAWBTemp, WBGTA, TWL, MD = readdata(A)
TD, WS, CwSB, HwSB, TempB, GTemp, WC, RH, HSI, DP, PWBTempB, SP, BP, Alt, DAlt, NAWBTemp, WBGTB, TWL, MD = readdata(B)
TD, WS, CwSC, HwSC, TempC, GTemp, WC, RH, HSI, DP, PWBTempC, SP, BP, Alt, DAlt, NAWBTemp, WBGTC, TWL, MD = readdata(C)
TD, WS, CwSD, HwSD, TempD, GTemp, WC, RH, HSI, DP, PWBTempD, SP, BP, Alt, DAlt, NAWBTemp, WBGTD, TWL, MD = readdata(D)
TD, WS, CwSE, HwSE, TempE, GTemp, WC, RH, HSI, DP, PWBTempE, SP, BP, Alt, DAlt, NAWBTemp, WBGTE, TWL, MD = readdata(E)
if vari == 'T':
D = [TempA,TempB,TempC,TempD,TempE]
elif vari == 'WBGT':
D = [WBGTA,WBGTB,WBGTC,WBGTD,WBGTE]
elif vari == 'CwS':
D = [CwSA,CwSB,CwSC,CwSD,CwSD,CwSE]
fig = plt.figure(figsize=(14,5))
ax1 = plt.subplot(251)
ax2 = plt.subplot(252)
ax3 = plt.subplot(253)
ax4 = plt.subplot(254)
ax5 = plt.subplot(255)
ax6 = plt.subplot(256)
ax7 = plt.subplot(257)
ax8 = plt.subplot(258)
ax9 = plt.subplot(259)
ax10 = plt.subplot(2,5,10)
axes = [ax1,ax2,ax3,ax4,ax5,ax6,ax7,ax8,ax9,ax10]
X = ['A','B','C','D','E']
a = 0
if vari == 'T':
vname = 'Temperature'
elif vari == 'WBGT':
vname = 'WBGT'
elif vari == 'CwS':
vname = 'Crosswind Speed'
#fig.suptitle('Scatter plots of ' + vname + ' values between sensors')
for i in range(5):
for j in range(i+1,5):
dat = pScatter(D[i],D[j])
axes[a].scatter(dat,D[j],c='b',s=0.2)
axes[a].set_xlabel(vname + "(Sensor "+X[i]+")")
axes[a].set_ylabel(vname + "(Sensor " + X[j] + ")")
a = a+1
plt.tight_layout()
plt.show()
def calcCi(A, B, C, D, E, vari):
TD, WSA, CwS, HwS, TempA, GTemp, WC, RH, HSI, DP, PWBTemp, SP, BP, Alt, DAlt, NAWBTemp, WBGT, TWL, MD = readdata(
A)
TD, WSB, CwS, HwS, TempB, GTemp, WC, RH, HSI, DP, PWBTemp, SP, BP, Alt, DAlt, NAWBTemp, WBGT, TWL, MD = readdata(
B)
TD, WSC, CwS, HwS, TempC, GTemp, WC, RH, HSI, DP, PWBTemp, SP, BP, Alt, DAlt, NAWBTemp, WBGT, TWL, MD = readdata(
C)
TD, WSD, CwS, HwS, TempD, GTemp, WC, RH, HSI, DP, PWBTemp, SP, BP, Alt, DAlt, NAWBTemp, WBGT, TWL, MD = readdata(
D)
TD, WSE, CwS, HwS, TempE, GTemp, WC, RH, HSI, DP, PWBTemp, SP, BP, Alt, DAlt, NAWBTemp, WBGT, TWL, MD = readdata(
E)
if vari == 'T':
A = [TempA, TempB, TempC, TempD, TempE]
print('For Temperature:')
if vari == 'WS':
A = [WSA, WSB, WSC, WSD, WSE]
print('For Windspeed:')
sensor_list = ['A', 'B', 'C', 'D', 'E']
lower_list = []
upper_list = []
for i in range(5):
mu, sigma = np.mean(A[i]), np.std(A[i])
size = len(A[i])
lower, upper = scipy.stats.t.interval(0.95, size - 1, mu, sigma)
lower_list.append(lower)
upper_list.append(upper)
intervals_list = pd.DataFrame([sensor_list, lower_list,
upper_list]).transpose()
intervals_list.columns = ['Sensor', 'Lower boundary', 'Upper boundary']
print(intervals_list)
def showCorrelation(A,B,C,D,E,vari):
TD, WS, CwSA, HwSA, TempA, GTemp, WC, RH, HSI, DP, PWBTempA, SP, BP, Alt, DAlt, NAWBTemp, WBGTA, TWL, MD = readdata(A)
TD, WS, CwSB, HwSB, TempB, GTemp, WC, RH, HSI, DP, PWBTempB, SP, BP, Alt, DAlt, NAWBTemp, WBGTB, TWL, MD = readdata(B)
TD, WS, CwSC, HwSC, TempC, GTemp, WC, RH, HSI, DP, PWBTempC, SP, BP, Alt, DAlt, NAWBTemp, WBGTC, TWL, MD = readdata(C)
TD, WS, CwSD, HwSD, TempD, GTemp, WC, RH, HSI, DP, PWBTempD, SP, BP, Alt, DAlt, NAWBTemp, WBGTD, TWL, MD = readdata(D)
TD, WS, CwSE, HwSE, TempE, GTemp, WC, RH, HSI, DP, PWBTempE, SP, BP, Alt, DAlt, NAWBTemp, WBGTE, TWL, MD = readdata(E)
if vari == 'T':
D = [TempA,TempB,TempC,TempD,TempE]
elif vari == 'WBGT':
D = [WBGTA,WBGTB,WBGTC,WBGTD,WBGTE]
elif vari == 'CwS':
D = [CwSA,CwSB,CwSC,CwSD,CwSD,CwSE]
a = np.zeros((5,5))
b = np.zeros((5,5))
c = np.zeros((5,5))
d = np.zeros((5,5))
for i in range(5):
for j in range(5):
a[i][j],b[i][j],c[i][j],d[i][j] = calcCorrelation(D[i],D[j])
if vari == 'T':
B = 'Temperature'
elif vari == 'WBGT':
B = 'Wet Bulb Globe Temperature'
elif vari == 'CwS':
B = 'Crosswind Speed'
print("\nPearson's coefficients for",B)
print(a)
print("\nSpearmann's coefficients for",B)
print(b)
def plotCdf(A, B, C, D, E, vari, n):
TD, WSA, CwS, HwS, TempA, GTemp, WC, RH, HSI, DP, PWBTemp, SP, BP, Alt, DAlt, NAWBTemp, WBGT, TWL, MD = readdata(
A)
TD, WSB, CwS, HwS, TempB, GTemp, WC, RH, HSI, DP, PWBTemp, SP, BP, Alt, DAlt, NAWBTemp, WBGT, TWL, MD = readdata(
B)
TD, WSC, CwS, HwS, TempC, GTemp, WC, RH, HSI, DP, PWBTemp, SP, BP, Alt, DAlt, NAWBTemp, WBGT, TWL, MD = readdata(
C)
TD, WSD, CwS, HwS, TempD, GTemp, WC, RH, HSI, DP, PWBTemp, SP, BP, Alt, DAlt, NAWBTemp, WBGT, TWL, MD = readdata(
D)
TD, WSE, CwS, HwS, TempE, GTemp, WC, RH, HSI, DP, PWBTemp, SP, BP, Alt, DAlt, NAWBTemp, WBGT, TWL, MD = readdata(
E)
fig = plt.figure(figsize=(14, 5))
ax1 = fig.add_subplot(111)
if vari == 'T':
a1 = ax1.hist(x=TempA,
bins=n,
cumulative=1,
density=1,
color='w',
alpha=0.7,
rwidth=0.85,
histtype='barstacked')
a2 = ax1.hist(x=TempB,
bins=n,
cumulative=1,
density=1,
color='w',
alpha=0.7,
rwidth=0.85,
histtype='barstacked')
a3 = ax1.hist(x=TempC,
bins=n,
cumulative=1,
density=1,
color='w',
alpha=0.7,
rwidth=0.85,
histtype='barstacked')
a4 = ax1.hist(x=TempD,
bins=n,
cumulative=1,
density=1,
color='w',
alpha=0.7,
rwidth=0.85,
histtype='barstacked')
a5 = ax1.hist(x=TempE,
bins=n,
cumulative=1,
density=1,
color='w',
alpha=0.7,
rwidth=0.85,
histtype='barstacked')
if vari == 'WS':
a1 = ax1.hist(x=WSA,
bins=n,
cumulative=1,
density=1,
color='w',
alpha=0.7,
rwidth=0.85,
histtype='barstacked')
a2 = ax1.hist(x=WSB,
bins=n,
cumulative=1,
density=1,
color='w',
alpha=0.7,
rwidth=0.85,
histtype='barstacked')
a3 = ax1.hist(x=WSC,
bins=n,
cumulative=1,
density=1,
color='w',
alpha=0.7,
rwidth=0.85,
histtype='barstacked')
a4 = ax1.hist(x=WSD,
bins=n,
cumulative=1,
density=1,
color='w',
alpha=0.7,
rwidth=0.85,
histtype='barstacked')
a5 = ax1.hist(x=WSE,
bins=n,
cumulative=1,
density=1,
color='w',
alpha=0.7,
rwidth=0.85,
histtype='barstacked')
ax1.plot(a1[1][1:] - (a1[1][1:] - a1[1][:-1]) / 2, a1[0], label='Sensor A')
ax1.plot(a2[1][1:] - (a2[1][1:] - a2[1][:-1]) / 2, a2[0], label='Sensor B')
ax1.plot(a3[1][1:] - (a3[1][1:] - a3[1][:-1]) / 2, a3[0], label='Sensor C')
ax1.plot(a4[1][1:] - (a4[1][1:] - a4[1][:-1]) / 2, a4[0], label='Sensor D')
ax1.plot(a5[1][1:] - (a5[1][1:] - a5[1][:-1]) / 2, a5[0], label='Sensor E')
plt.ylabel('Cumulative Density')
if vari == 'T':
plt.xlabel('Temperature [\N{DEGREE SIGN}C]')
#plt.title('CDFs for the 5 sensors Temperature values')
if vari == 'WS':
plt.xlabel('Windspeed [m/s]')
#plt.title('CDFs for the 5 sensors Windspeed values')
plt.legend(loc='best')
plt.tight_layout()
plt.show()
def plotWSPdfkde(A, B, C, D, E, n):
TD, WSA, CwS, HwS, TempA, GTemp, WC, RH, HSI, DP, PWBTemp, SP, BP, Alt, DAlt, NAWBTemp, WBGT, TWL, MD = readdata(
A)
TD, WSB, CwS, HwS, TempB, GTemp, WC, RH, HSI, DP, PWBTemp, SP, BP, Alt, DAlt, NAWBTemp, WBGT, TWL, MD = readdata(
B)
TD, WSC, CwS, HwS, TempC, GTemp, WC, RH, HSI, DP, PWBTemp, SP, BP, Alt, DAlt, NAWBTemp, WBGT, TWL, MD = readdata(
C)
TD, WSD, CwS, HwS, TempD, GTemp, WC, RH, HSI, DP, PWBTemp, SP, BP, Alt, DAlt, NAWBTemp, WBGT, TWL, MD = readdata(
D)
TD, WSE, CwS, HwS, TempE, GTemp, WC, RH, HSI, DP, PWBTemp, SP, BP, Alt, DAlt, NAWBTemp, WBGT, TWL, MD = readdata(
E)
ax1 = plt.subplot(231)
ax2 = plt.subplot(232)
ax3 = plt.subplot(233)
ax4 = plt.subplot(234)
ax5 = plt.subplot(235)
ax1.hist(x=WSA, bins=n, density=True, histtype='step', label='PDF')
ax1.set_xlabel('Sensor A')
ax2.hist(x=WSB, bins=n, density=True, histtype='step', label='PDF')
ax2.set_xlabel('Sensor B')
ax3.hist(x=WSC, bins=n, density=True, histtype='step', label='PDF')
ax3.set_xlabel('Sensor C')
ax4.hist(x=WSD, bins=n, density=True, histtype='step', label='PDF')
ax4.set_xlabel('Sensor D')
ax5.hist(x=WSE, bins=n, density=True, histtype='step', label='PDF')
ax5.set_xlabel('Sensor E')
sns.kdeplot(WSA, color='springgreen', ax=ax1, label="KDE")
sns.kdeplot(WSB, color='springgreen', ax=ax2, label="KDE")
sns.kdeplot(WSC, color='springgreen', ax=ax3, label="KDE")
sns.kdeplot(WSD, color='springgreen', ax=ax4, label="KDE")
sns.kdeplot(WSE, color='springgreen', ax=ax5, label="KDE")
plt.legend(loc='lower left', bbox_to_anchor=(1.5, 0))
#plt.suptitle('PDFs and the kernel density estimation for the 5 sensors Wind Speed values', )
plt.subplots_adjust(wspace=0.3)
plt.subplots_adjust(hspace=0.5)
plt.show()
def plotPdf(A, B, C, D, E):
TD, WS, CwS, HwS, TempA, GTemp, WC, RH, HSI, DP, PWBTemp, SP, BP, Alt, DAlt, NAWBTemp, WBGT, TWL, MD = readdata(
A)
TD, WS, CwS, HwS, TempB, GTemp, WC, RH, HSI, DP, PWBTemp, SP, BP, Alt, DAlt, NAWBTemp, WBGT, TWL, MD = readdata(
B)
TD, WS, CwS, HwS, TempC, GTemp, WC, RH, HSI, DP, PWBTemp, SP, BP, Alt, DAlt, NAWBTemp, WBGT, TWL, MD = readdata(
C)
TD, WS, CwS, HwS, TempD, GTemp, WC, RH, HSI, DP, PWBTemp, SP, BP, Alt, DAlt, NAWBTemp, WBGT, TWL, MD = readdata(
D)
TD, WS, CwS, HwS, TempE, GTemp, WC, RH, HSI, DP, PWBTemp, SP, BP, Alt, DAlt, NAWBTemp, WBGT, TWL, MD = readdata(
E)
mu1, sigma1 = np.mean(TempA), np.std(TempA)
mu2, sigma2 = np.mean(TempB), np.std(TempB)
mu3, sigma3 = np.mean(TempC), np.std(TempC)
mu4, sigma4 = np.mean(TempD), np.std(TempD)
mu5, sigma5 = np.mean(TempE), np.std(TempE)
x1 = np.linspace(mu1 - 4 * sigma1, mu1 + 4 * sigma1, 100)
x2 = np.linspace(mu2 - 4 * sigma2, mu2 + 4 * sigma2, 100)
x3 = np.linspace(mu3 - 4 * sigma3, mu3 + 4 * sigma3, 100)
x4 = np.linspace(mu4 - 4 * sigma4, mu4 + 4 * sigma4, 100)
x5 = np.linspace(mu5 - 4 * sigma5, mu5 + 4 * sigma5, 100)
plt.figure(figsize=(14, 5))
plt.plot(x1, stats.norm.pdf(x1, mu1, sigma1), label='Sensor A')
plt.plot(x2, stats.norm.pdf(x2, mu2, sigma2), label='Sensor B')
plt.plot(x3, stats.norm.pdf(x3, mu3, sigma3), label='Sensor C')
plt.plot(x4, stats.norm.pdf(x4, mu4, sigma4), label='Sensor D')
plt.plot(x5, stats.norm.pdf(x5, mu5, sigma5), label='Sensor E')
plt.ylabel('Probability Density')
plt.xlabel('Temperature [\N{DEGREE SIGN}C]')
#plt.title('PDFs for the 5 sensors Temperature values')
plt.ylim(ymin=0)
plt.legend(loc='best')
plt.show()
def plotPmf(A, B, C, D, E, n):
TD, WS, CwS, HwS, TempA, GTemp, WC, RH, HSI, DP, PWBTemp, SP, BP, Alt, DAlt, NAWBTemp, WBGT, TWL, MD = readdata(
A)
TD, WS, CwS, HwS, TempB, GTemp, WC, RH, HSI, DP, PWBTemp, SP, BP, Alt, DAlt, NAWBTemp, WBGT, TWL, MD = readdata(
B)
TD, WS, CwS, HwS, TempC, GTemp, WC, RH, HSI, DP, PWBTemp, SP, BP, Alt, DAlt, NAWBTemp, WBGT, TWL, MD = readdata(
C)
TD, WS, CwS, HwS, TempD, GTemp, WC, RH, HSI, DP, PWBTemp, SP, BP, Alt, DAlt, NAWBTemp, WBGT, TWL, MD = readdata(
D)
TD, WS, CwS, HwS, TempE, GTemp, WC, RH, HSI, DP, PWBTemp, SP, BP, Alt, DAlt, NAWBTemp, WBGT, TWL, MD = readdata(
E)
fig = plt.figure(figsize=(14, 5))
ax1 = fig.add_subplot(231)
ax2 = fig.add_subplot(232)
ax3 = fig.add_subplot(233)
ax4 = fig.add_subplot(234)
ax5 = fig.add_subplot(235)
weights = np.ones_like(TempE) / float(len(TempE))
ax1.hist(x=TempA, bins=n, density=True, color='b', histtype='step')
ax1.set_ylabel('Probability')
ax1.set_xlabel('Temperature [\N{DEGREE SIGN}C]\nSensor A')
ax2.hist(x=TempB, bins=n, density=True, color='b', histtype='step')
ax2.set_ylabel('Probability')
ax2.set_xlabel('Temperature [\N{DEGREE SIGN}C]\nSensor B')
ax3.hist(x=TempC, bins=n, density=True, color='b', histtype='step')
ax3.set_ylabel('Probability')
ax3.set_xlabel('Temperature [\N{DEGREE SIGN}C]\nSensor C')
ax4.hist(x=TempD, bins=n, density=True, color='b', histtype='step')
ax4.set_ylabel('Probability')
ax4.set_xlabel('Temperature [\N{DEGREE SIGN}C]\nSensor D')
ax5.hist(x=TempE,
bins=n,
density=True,
color='b',
histtype='step',
weights=weights)
ax5.set_ylabel('Probability')
ax5.set_xlabel('Temperature [\N{DEGREE SIGN}C]\nSensor E')
#fig.suptitle('PMFs for the 5 sensors Temperature values',)
plt.subplots_adjust(wspace=0.5)
plt.subplots_adjust(hspace=0.5)
plt.show()