基于机器学习的变电站信息清洗与重构的研究

目前采用人工对信号进行判断、识别、梳理。但人为的梳理结果不甚理想,进度也极为缓慢,且依赖人员的专业水平和经验,同时人工识别的正确率难以保证,容易出现误判、疏漏。因此,亟需采取人工智能领域相关技术,遵循依托南方电网调度与自动化专业领域标准规范和业务要求,构建一套监控信息的数据清洗与重构,构建变电站信息清洗与重构机器学习模型,提高信号识别匹配能力、降低安全技术风险、减轻人员工作负担,为智能告警的实现提供准备,全面提升电网的监控效率。

Research on Cleaning and Reconstruction of Substation Information Based on Machine Learning

At present, manual signal is used to judge, identify and comb. But the artificial carding result is not very ideal, the progress is also extremely slow, and depends on the personnel’s professional level and experience, at the same time, the correct rate of manual recognition is difficult to guarantee, prone to misjudgment, omission. Therefore, it is urgent to adopt relevant technologies in the field of artificial intelligence, to construct a set of data cleaning and reconstruction of monitoring information, and to construct a machine learning model for substation information cleaning and reconstruction. Improve signal recognition and matching ability, reduce the risk of safety technology, reduce the workload of personnel for intelligent alarm.And the monitoring efficiency of the power grid is comprehensively improved.