南京信息工程大学学报2025，Vol.17Issue(3)：384-395,12.DOI:10.13878/j.cnki.jnuist.20240508003
基于SCADA图像数据增强的风机叶片结冰检测
Detection of wind turbine blade icing based on SCADA image data enhancement
摘要
Abstract
Here,a SCADA(Supervisory Control And Data Acquisition)image data enhancement method is pro-posed to detect wind turbine blade icing.The method initially reconstructs the original features,and features are se-lected by considering both data relevance and feature importance.Subsequently,the SCADA data is transformed into a two-dimensional image format to meet the input requirements of two-dimensional neural network models.On this basis,an optimized CycleGAN algorithm is used to generate more adaptable image data,aiming to address the issue of imbalanced data category and significantly enhance the generalization ability of the model.The WT15 data is used as the training set and WT21 data as the test set.The experimental results demonstrate that the proposed feature se-lection method achieves an increase of 4.69 percentage points in model accuracy and 2.64 percentage points in F1 score,compared to feature selection using the XGBoost model.In comparison to the original CycleGAN model,the improved model achieved an increase in accuracy of 6.78 percentage points and an improvement in the F1 score of 6.91 percentage points.These findings indicate that the proposed approach offers a significant advantage in improving the model's accuracy and generalization ability.关键词
叶片结冰/异常检测/数据图形化/XGBoost/CycleGANKey words
blade icing/anomaly detection/data visualization/XGBoost/CycleGAN分类
动力与电气工程引用本文复制引用
王昊,姚刚,卢应强,葛泉波..基于SCADA图像数据增强的风机叶片结冰检测[J].南京信息工程大学学报,2025,17(3):384-395,12.基金项目
江苏高校"青蓝工程"项目(R2023Q07) （R2023Q07）
