基于跨域少样本学习的电网作业违章动作分类

Illegal Action Classification in Power Grid Operation Based on Cross-domain Few-shot Learning

为了实现电网作业智能监控中的违章动作分类,以提高电力系统运维的效率和安全性,减少对样本标注的依赖,该文提出了一种基于跨域少样本学习的电网作业违章动作分类方法。该方法设计了一种创新的跨域对齐机制,通过构建域间生成机制和域内扩展机制生成跨域辅助数据集和目标域扩展数据集,帮助分类模型更好地理解和适应不同域之间的特征变化,以及增强模型在目标域中的不变性学习能力,从而提高在电网作业场景中不同客观因素下违章动作分类的准确性和效率。实验结果表明,提出的方法有效地降低了对大规模标注数据的依赖性,通过跨域少样本学习方法,将未知的无标注样本数据输入分类模型,实现了对电网作业场景中违章动作的高效准确分类,展现出在实际电网运维中的广泛应用前景。

In order to realize the classification of illegal actions in intelligent monitoring of grid operations,improve the efficiency and safety of power system operation and maintenance,and reduce the dependence on sample labeling,this paper proposes a method of illegal actions classification of power grid operations based on cross-domain few-shot learning.In this method,an innovative cross-domain alignment mechanism is designed to generate cross-domain auxiliary data sets and target domain extended data sets by constructing an inter-domain generation mechanism and an intra-domain extension mechanism,which helps the classification model to better understand and adapt to the feature changes between different domains,and enhances the model's invariance learning ability in the target domain,so as to improve the accuracy and efficiency of the classification of illegal actions under different objective factors in the power grid operation scenario.Experimental results show that the the proposed method effectively reduces the dependence on large-scale annotated data,and through the cross-domain few-shot learning method,the unknown unlabeled sample data are input into the classification model,and the efficient and accurate classification of illegal actions in the power grid operation scenario is realized,showing a wide application prospect in the actual power grid operation and maintenance.