Wind Turbine Blade Icing Prediction Based on Deep Belief Network

2019 4th International Conference on Mechanical, Control and Computer Engineering (ICMCCE)(2019)

Cited 7|Views0
No score
Abstract
Aiming at the problem that wind turbine blade icing poses a great threat to the power generation performance and safe operation of wind turbine, a blade icing prediction method based on deep belief network (DBN) is proposed in this paper. Firstly, the invalid data in the original data are removed, and preprocessing such as labeling, averaging and normalization is performed. The resampling method is used to solve the problem of uneven distribution of positive and negative samples. According to the relationship model between icing and wind turbine performance, the features of wind speed and power related to icing fault are selected. Secondly, a four-layer DBN network consisting of two RBMs and one classifier is established. After the model parameters are set, the features and labels of the training set are placed in the DBN for training. Get the optimal model after multiple iterations. Finally, the test set is predicted and evaluated. Compared with SVM algorithm, this method has high prediction accuracy and strong stability, and has a significant effect on improving the efficiency of deicing system and reducing the risk of wind turbine operation.
More
Translated text
Key words
wind turbine blades,icing prediction,unbalanced samples,DBN
AI Read Science
Must-Reading Tree
Example
Generate MRT to find the research sequence of this paper
Chat Paper
Summary is being generated by the instructions you defined