基于机器学习的智能化电力设备巡检技术应用

Application of Intelligent Power Equipment Inspection Technology Based on Machine Learning

阐述数据的采集和预处理方法,通过支持向量机和决策树等机器学习算法进行模型训练,实现对设备状态的智能诊断和异常检测,并以数据可视化方式进行展示。针对巡检过程中出现的异常情况,提出基于神经网络的异常分类和定位方案。

This paper describes the methods of data collection and preprocessing,using machine learning algorithms such as support vector machines and decision trees for model training,to achieve intelligent diagnosis and anomaly detection of device status,and display it in a data visualization manner.It proposes a neural network-based anomaly classification and localization scheme for the abnormal situations that occur during the inspection process.