完成--预测风电功率测试初步完成
This commit is contained in:
parent
92bc054265
commit
1a8cbb5f06
@ -44,10 +44,54 @@
|
||||
(1)基于天气与历史数据预测风力发电量
|
||||
(2)XGBoost,LSTM,Prophet时间序列模型
|
||||
(3)Wind Turbine Scada Dataset
|
||||
https://www.kaggle.com/datasets/berkerisen/wind-turbine-scada-dataset/discussion/520518
|
||||
|
||||
数据集解析:
|
||||
数据集路径:dataset/dataset_raw/WindTurbine/wind_turbine_scada.csv
|
||||
列:
|
||||
Date/Time,LV ActivePower (kW),Wind Speed (m/s),Theoretical_Power_Curve (KWh),Wind Direction (°)
|
||||
注: LV ActivePower 为实际电机功率, Theoretical_Power_Curve为理论上的电机功率.
|
||||
只有 Wind Speed 和 Wind Direction 有用. LV ActivePower 作为数据真值使用.
|
||||
|
||||
数据集处理:
|
||||
重命名列, 将真值放到最后,去除 date和 theoretical 两列.
|
||||
去除真值为0的行.
|
||||
wind_0.csv
|
||||
|
||||
标准缩放--结果分析:
|
||||
RandomForestRegressor:mae 123.
|
||||
XGBRegressor:mae 116
|
||||
LGBMRegressor: mae 113
|
||||
AdaBoostRegressor mae:277
|
||||
不进行缩放--结果分析
|
||||
RandomForestRegressor:mae 123.861827
|
||||
XGBRegressor:mae 116.247
|
||||
LGBMRegressor: mae 112.53
|
||||
AdaBoostRegressor mae:217.83
|
||||
|
||||
在上述处理基础上,计算相邻两天风速的差值gust,把gust作为新的特征列.
|
||||
wind_1.csv
|
||||
|
||||
标准缩放--结果分析:
|
||||
RandomForestRegressor:mae 122.
|
||||
XGBRegressor:mae 121.9
|
||||
LGBMRegressor: mae 115.50
|
||||
AdaBoostRegressor mae:321.77
|
||||
原始数据--结果分析:
|
||||
RandomForestRegressor:mae 122.64
|
||||
XGBRegressor:mae 121.93
|
||||
LGBMRegressor: mae 115.50
|
||||
AdaBoostRegressor mae:253.43
|
||||
|
||||
|
||||
|
||||
|
||||
|
||||
3.化工行业-过程异常检测
|
||||
(1)分类化工生产过程中的异常状态(泄露,温度失控)
|
||||
(2)SVM,孤立森林,Autoencoder
|
||||
(3)ennessee Eastman Process (TEP):模拟化工厂故障的多变量时序数据。
|
||||
(3)Tennessee Eastman Process (TEP):模拟化工厂故障的多变量时序数据。
|
||||
https://github.com/camaramm/tennessee-eastman-profBraatz
|
||||
4.物流与供应链-需求预测
|
||||
(1)预测未来产品需求量以优化库存
|
||||
(2)ARIMA,LightGBM, Transformer
|
||||
|
||||
Binary file not shown.
Loading…
Reference in New Issue
Block a user