Building cold load prediction method based on LightGBM under machine learning framework
The invention discloses a building cold load prediction method based on LightGBM under a machine learning framework, and the method comprises the steps: selecting outdoor meteorological data and indoor environment data in a given time period, and constructing a data set; preprocessing the data of th...
Saved in:
Main Authors | , , , , |
---|---|
Format | Patent |
Language | Chinese English |
Published |
31.07.2020
|
Subjects | |
Online Access | Get full text |
Cover
Loading…
Summary: | The invention discloses a building cold load prediction method based on LightGBM under a machine learning framework, and the method comprises the steps: selecting outdoor meteorological data and indoor environment data in a given time period, and constructing a data set; preprocessing the data of the data set, wherein the preprocessing specifically comprises data cleaning, correlation analysis andstandardization processing; dividing the preprocessed data set into a training set, a verification set and a test set; importing the LightGBM model and setting model parameters; loading the data setpreprocessed in the step 2 into a Data object, training a LightGBM model, and setting training parameters; predicting the cold load, and outputting a predicted building cold load value. Compared withother prediction models, the method has the advantages that the calculation efficiency and the prediction accuracy are remarkably improved, and the building cold load prediction efficiency and precision are improved.
本发明公开了一种基于 |
---|---|
Bibliography: | Application Number: CN202010279924 |