Python Regression Program of Combined Cycle Power Plant

In [1]: import numpy as np import pandas as pd import matplotlib.pyplot as plt import seaborn as sns from sklearn.l...

In [1]:

import numpy as np
import pandas as pd
import matplotlib.pyplot as plt
import seaborn as sns
from sklearn.linear_model import LinearRegression
from sklearn.preprocessing import PolynomialFeatures

In [2]:

df = pd.read_excel("Folds5x2_pp.xlsx")

In [3]:

df

Out[3]:

	AT	V	AP	RH	PE
0	14.96	41.76	1024.07	73.17	463.26
1	25.18	62.96	1020.04	59.08	444.37
2	5.11	39.40	1012.16	92.14	488.56
3	20.86	57.32	1010.24	76.64	446.48
4	10.82	37.50	1009.23	96.62	473.90
5	26.27	59.44	1012.23	58.77	443.67
6	15.89	43.96	1014.02	75.24	467.35
7	9.48	44.71	1019.12	66.43	478.42
8	14.64	45.00	1021.78	41.25	475.98
9	11.74	43.56	1015.14	70.72	477.50
10	17.99	43.72	1008.64	75.04	453.02
11	20.14	46.93	1014.66	64.22	453.99
12	24.34	73.50	1011.31	84.15	440.29
13	25.71	58.59	1012.77	61.83	451.28
14	26.19	69.34	1009.48	87.59	433.99
15	21.42	43.79	1015.76	43.08	462.19
16	18.21	45.00	1022.86	48.84	467.54
17	11.04	41.74	1022.60	77.51	477.20
18	14.45	52.75	1023.97	63.59	459.85
19	13.97	38.47	1015.15	55.28	464.30
20	17.76	42.42	1009.09	66.26	468.27
21	5.41	40.07	1019.16	64.77	495.24
22	7.76	42.28	1008.52	83.31	483.80
23	27.23	63.90	1014.30	47.19	443.61
24	27.36	48.60	1003.18	54.93	436.06
25	27.47	70.72	1009.97	74.62	443.25
26	14.60	39.31	1011.11	72.52	464.16
27	7.91	39.96	1023.57	88.44	475.52
28	5.81	35.79	1012.14	92.28	484.41
29	30.53	65.18	1012.69	41.85	437.89
...	...	...	...	...	...
9538	8.64	38.56	1016.51	66.03	484.45
9539	10.53	37.50	1008.55	99.91	472.32
9540	23.53	50.05	1005.63	78.40	443.71
9541	24.90	67.25	1017.77	66.17	433.71
9542	5.01	39.40	1003.69	91.90	475.34
9543	22.66	69.84	1006.16	82.79	439.06
9544	29.76	57.19	1008.59	51.10	436.21
9545	26.30	61.41	1012.45	56.85	448.55
9546	30.17	74.22	1007.46	49.27	432.00
9547	8.02	40.23	1017.42	90.26	484.22
9548	19.12	50.16	1011.52	99.71	451.49
9549	14.87	42.18	1015.23	74.41	465.89
9550	9.71	42.44	1014.29	94.03	481.03
9551	24.33	77.54	1008.50	82.45	435.38
9552	7.17	39.40	1011.48	90.38	484.33
9553	24.61	62.96	1020.10	63.83	445.79
9554	23.48	66.44	1011.28	61.11	443.21
9555	23.70	70.32	1007.21	66.85	439.59
9556	25.44	69.59	1008.22	80.73	433.97
9557	17.46	62.10	1019.96	83.99	451.06
9558	22.97	62.40	1010.25	75.18	445.30
9559	26.22	49.82	1015.48	55.80	454.20
9560	23.27	68.28	1005.01	74.83	444.86
9561	11.76	41.58	1020.91	88.35	465.45
9562	14.02	40.10	1015.56	82.44	467.32
9563	16.65	49.69	1014.01	91.00	460.03
9564	13.19	39.18	1023.67	66.78	469.62
9565	31.32	74.33	1012.92	36.48	429.57
9566	24.48	69.45	1013.86	62.39	435.74
9567	21.60	62.52	1017.23	67.87	453.28

9568 rows × 5 columns

In [4]:

np.mean(df)

Out[4]:

AT      19.651231
V       54.305804
AP    1013.259078
RH      73.308978
PE     454.365009
dtype: float64

In [5]:

df.median()

Out[5]:

AT      20.345
V       52.080
AP    1012.940
RH      74.975
PE     451.550
dtype: float64

In [6]:

df.std()

Out[6]:

AT     7.452473
V     12.707893
AP     5.938784
RH    14.600269
PE    17.066995
dtype: float64

In [7]:

df.corr()

Out[7]:

	AT	V	AP	RH	PE
AT	1.000000	0.844107	-0.507549	-0.542535	-0.948128
V	0.844107	1.000000	-0.413502	-0.312187	-0.869780
AP	-0.507549	-0.413502	1.000000	0.099574	0.518429
RH	-0.542535	-0.312187	0.099574	1.000000	0.389794
PE	-0.948128	-0.869780	0.518429	0.389794	1.000000

In [8]:

sns.heatmap(df.corr(), annot = True)
plt.show()

In [9]:

ambient_temperature = df['AT']
exhaust_vacuum = df['V']
ambient_pressure = df['AP']
relative_humidity = df['RH']
plant_energy_output = df['PE']

In [10]:

fig,ccpp = plt.subplots(2,3)

In [11]:

ccpp[0,0].boxplot(ambient_temperature)
ccpp[0,1].boxplot(exhaust_vacuum)
ccpp[0,2].boxplot(ambient_pressure)
ccpp[1,0].boxplot(relative_humidity)
ccpp[1,1].boxplot(plant_energy_output)

Out[11]:

{'boxes': [<matplotlib.lines.Line2D at 0x7f17330399b0>],
 'caps': [<matplotlib.lines.Line2D at 0x7f1733041400>,
  <matplotlib.lines.Line2D at 0x7f1733041828>],
 'fliers': [<matplotlib.lines.Line2D at 0x7f173304a0b8>],
 'means': [],
 'medians': [<matplotlib.lines.Line2D at 0x7f1733041c50>],
 'whiskers': [<matplotlib.lines.Line2D at 0x7f1733039b00>,
  <matplotlib.lines.Line2D at 0x7f1733039f98>]}

In [12]:

plt.figure(figsize=(20,8))
plt.show()

<matplotlib.figure.Figure at 0x7f173306cba8>

In [13]:

#ambient_temperature

In [14]:

plt.hist(ambient_temperature,bins=10)
plt.show()

In [15]:

#ambient_pressure

In [16]:

plt.hist(ambient_pressure,bins=20)
plt.show()

In [17]:

plt.hist(exhaust_vacuum,bins=10)
plt.show()

In [18]:

plt.hist(relative_humidity,bins=10)
plt.show()

In [19]:

plt.hist(plant_energy_output,bins=20)
plt.show()

In [20]:

#80% data as Train dataset and 20% as Test dataset

In [21]:

xtrain = df.iloc[:7655,:4]
ytrain = df.iloc[:7655,4:]
#xtrain

In [22]:

xtest = df.iloc[7655:,:4]
ytest = df.iloc[7655:,4:]
#ytest

In [23]:

lin_model = LinearRegression()

In [24]:

lin_model.fit(xtrain,ytrain)

Out[24]:

LinearRegression(copy_X=True, fit_intercept=True, n_jobs=1, normalize=False)

In [25]:

lin_model.predict(xtest)

Out[25]:

array([[441.82904193],
       [445.14394502],
       [426.15474298],
       ...,
       [432.47274932],
       [443.10509547],
       [449.74518291]])

In [26]:

lin_model.score(xtest,ytest)

Out[26]:

0.9246449943523912

In [27]:

sns.lmplot(x="AT", y="PE", data=df)
plt.title('Ambident Temerature and Plant Energy Relation')
plt.show()

In [28]:

sns.lmplot(x="AP", y="PE", data=df)
plt.title('Ambident Pressure and Plant Energy Relation')
plt.show()

In [29]:

sns.lmplot(x="V", y="PE", data=df)
plt.title('Exhaust Volume and Plant Energy Relation')
plt.show()

In [30]:

sns.lmplot(x="RH", y="PE", data=df)
plt.title('Relative Humidity and Plant Energy Relation')
plt.show()

In [31]:

quad_feature = PolynomialFeatures(degree=2)
xtrain_quad = quad_feature.fit_transform(xtrain)
xtest_quad = quad_feature.transform(xtest)
quad_model = LinearRegression()
quad_model.fit(xtrain_quad,ytrain)
quad_model.predict(xtest_quad)
quad_model.score(xtest_quad,ytest)

Out[31]:

0.9338502280297939

In [32]:

quad_feature = PolynomialFeatures(degree=3)
xtrain_quad = quad_feature.fit_transform(xtrain)
xtest_quad = quad_feature.transform(xtest)
quad_model = LinearRegression()
quad_model.fit(xtrain_quad,ytrain)
quad_model.predict(xtest_quad)
quad_model.score(xtest_quad,ytest)

Out[32]:

0.9367778711840852

In [33]:

quad_feature = PolynomialFeatures(degree=4)
xtrain_quad = quad_feature.fit_transform(xtrain)
xtest_quad = quad_feature.transform(xtest)
quad_model = LinearRegression()
quad_model.fit(xtrain_quad,ytrain)
quad_model.predict(xtest_quad)
quad_model.score(xtest_quad,ytest)

Out[33]:

0.9380846117329315

In [34]:

quad_feature = PolynomialFeatures(degree=5)
xtrain_quad = quad_feature.fit_transform(xtrain)
xtest_quad = quad_feature.transform(xtest)
quad_model = LinearRegression()
quad_model.fit(xtrain_quad,ytrain)
quad_model.predict(xtest_quad)
quad_model.score(xtest_quad,ytest)

Out[34]:

0.936743890334244

In [35]:

quad_feature = PolynomialFeatures(degree=6)
xtrain_quad = quad_feature.fit_transform(xtrain)
xtest_quad = quad_feature.transform(xtest)
quad_model = LinearRegression()
quad_model.fit(xtrain_quad,ytrain)
quad_model.predict(xtest_quad)
quad_model.score(xtest_quad,ytest)

Out[35]:

0.9388071352817638

In [36]:

quad_feature = PolynomialFeatures(degree=7)
xtrain_quad = quad_feature.fit_transform(xtrain)
xtest_quad = quad_feature.transform(xtest)
quad_model = LinearRegression()
quad_model.fit(xtrain_quad,ytrain)
quad_model.predict(xtest_quad)
quad_model.score(xtest_quad,ytest)

Out[36]:

0.9364924482642509

In [37]:

quad_feature = PolynomialFeatures(degree=8)
xtrain_quad = quad_feature.fit_transform(xtrain)
xtest_quad = quad_feature.transform(xtest)
quad_model = LinearRegression()
quad_model.fit(xtrain_quad,ytrain)
quad_model.predict(xtest_quad)
quad_model.score(xtest_quad,ytest)

Out[37]:

0.9391750864551984

In [38]:

quad_feature = PolynomialFeatures(degree=9)
xtrain_quad = quad_feature.fit_transform(xtrain)
xtest_quad = quad_feature.transform(xtest)
quad_model = LinearRegression()
quad_model.fit(xtrain_quad,ytrain)
quad_model.predict(xtest_quad)
quad_model.score(xtest_quad,ytest)

Out[38]:

0.9380537936399923

In [39]:

quad_feature = PolynomialFeatures(degree=10)
xtrain_quad = quad_feature.fit_transform(xtrain)
xtest_quad = quad_feature.transform(xtest)
quad_model = LinearRegression()
quad_model.fit(xtrain_quad,ytrain)
quad_model.predict(xtest_quad)
quad_model.score(xtest_quad,ytest)

Out[39]:

0.9383741698139699

In [40]:

plt.scatter(xtrain['AT'],ytrain)
plt.show()

In [ ]:

Get dataset from "https://github.com/Jakesh-Bohaju/Python-Data-Science/blob/master/Folds5x2_pp.xlsx"

In [ ]: