ICEEMDAN-Solar_power-forecast/iceemdan信号重构.ipynb

In [18]:

from math import sqrt
from numpy import concatenate
from matplotlib import pyplot
import pandas as pd
import numpy as np
from sklearn.preprocessing import MinMaxScaler
from sklearn.preprocessing import LabelEncoder
from sklearn.metrics import mean_squared_error
from tensorflow.keras import Sequential

from tensorflow.keras.layers import Dense
from tensorflow.keras.layers import LSTM
from tensorflow.keras.layers import Dropout
from sklearn.model_selection import train_test_split
import matplotlib.pyplot as plt

In [19]:

# 加载数据
path1 = r"D:\project\小论文1-基于ICEEMDAN分解的时序高维变化的短期光伏功率预测模型\CEEMAN-PosConv1dbiLSTM-LSTM\模型代码流程\完整的模型代码流程\低频_forecast.csv"#数据所在路径
#我的数据是excel表，若是csv文件用pandas的read_csv()函数替换即可。
f_low= pd.DataFrame(pd.read_csv(path1))

In [20]:

# 加载数据
path2 = r"D:\project\小论文1-基于ICEEMDAN分解的时序高维变化的短期光伏功率预测模型\CEEMAN-PosConv1dbiLSTM-LSTM\模型代码流程\完整的模型代码流程\高频re_forecast.csv"#数据所在路径
#我的数据是excel表，若是csv文件用pandas的read_csv()函数替换即可。
f_high= pd.DataFrame(pd.read_csv(path2))

In [21]:

path3= r"D:\project\小论文1-基于ICEEMDAN分解的时序高维变化的短期光伏功率预测模型\CEEMAN-PosConv1dbiLSTM-LSTM\模型代码流程\完整的模型代码流程\低频_test.csv"#数据所在路径
#我的数据是excel表，若是csv文件用pandas的read_csv()函数替换即可。
true_low= pd.DataFrame(pd.read_csv(path3))

In [22]:

path4= r"D:\project\小论文1-基于ICEEMDAN分解的时序高维变化的短期光伏功率预测模型\CEEMAN-PosConv1dbiLSTM-LSTM\模型代码流程\完整的模型代码流程\高频re_test.csv"#数据所在路径
#我的数据是excel表，若是csv文件用pandas的read_csv()函数替换即可。
true_high= pd.DataFrame(pd.read_csv(path4))

In [23]:

pre_data=f_low+f_high
pre_data

Out[23]:

	column_name
0	1.958189
1	2.254070
2	1.279676
3	1.747101
4	1.987240
...	...
20826	0.017192
20827	0.014549
20828	0.014636
20829	0.016639
20830	0.017215

20831 rows × 1 columns

In [24]:

true=true_low+true_high
true

Out[24]:

	column_name
0	2.186333e+00
1	1.377467e+00
2	1.452000e+00
3	1.846867e+00
4	2.793334e+00
...	...
20826	6.661338e-16
20827	0.000000e+00
20828	2.220446e-16
20829	2.220446e-16
20830	4.440892e-16

20831 rows × 1 columns

In [25]:

plt.figure(figsize=(16,8))
plt.plot(true, label='true')
plt.plot(pre_data, label='pre')
plt.legend()
plt.show()

In [29]:

from sklearn.metrics import mean_squared_error, mean_absolute_error  # 评价指标
# 使用sklearn调用衡量线性回归的MSE 、 RMSE、 MAE、r2
from math import sqrt
from sklearn.metrics import mean_absolute_error
from sklearn.metrics import mean_squared_error
from sklearn.metrics import r2_score
print('mean_squared_error:', mean_squared_error(pre_data, true))  # mse)
print("mean_absolute_error:", mean_absolute_error(pre_data, true))  # mae
print("rmse:", sqrt(mean_squared_error(pre_data, true)))
print("r2 score:", r2_score(pre_data[5000:10000], true[5000:10000]))#预测50天

mean_squared_error: 0.04969353670622598
mean_absolute_error: 0.08076025073121713
rmse: 0.22292047170734675
r2 score: 0.9988271323117631

In [46]:

# 使用MinMaxScaler进行归一化
from sklearn.preprocessing import MinMaxScaler
scaler = MinMaxScaler(feature_range=(0, 1))
pre = scaler.fit_transform(pre_data)
print(pre.shape)

(20831, 1)

In [47]:

from sklearn.preprocessing import MinMaxScaler
scaler = MinMaxScaler(feature_range=(0, 1))
true_data = scaler.fit_transform(true)
print(true_data.shape)

(20831, 1)

In [50]:

from sklearn.metrics import mean_squared_error, mean_absolute_error  # 评价指标
# 使用sklearn调用衡量线性回归的MSE 、 RMSE、 MAE、r2
from math import sqrt
from sklearn.metrics import mean_absolute_error
from sklearn.metrics import mean_squared_error
from sklearn.metrics import r2_score
print('mean_squared_error:', mean_squared_error(pre, true_data))  # mse)
print("mean_absolute_error:", mean_absolute_error(pre, true_data))  # mae
print("rmse:", sqrt(mean_squared_error(pre, true_data)))
print("r2 score:", r2_score(pre_data, true))

mean_squared_error: 0.0026778377010073626
mean_absolute_error: 0.027468762691519367
rmse: 0.05174782798347543
r2 score: 0.9988074259067585