电力设备巡检缺陷图像智能识别技术研究

Research on Intelligent Image Identification Technology of Power Equipment Inspection Defects

电力设备具有点多、面广、长期暴露野外等特点,运行过程中易受外界环境等多方面因素影响。外界环境因素包括风偏、鸟害、覆冰、雷击、山火等,其他因素还有设备老化、质量、设计、验收、运维等,单一因素或多个综合因素都可导致设备故障,进而造成大面积停电,严重影响设备运行。随着无人机电力设备巡检的应用和发展,无人机巡检数据不断的积累,通过智能化处理手段,利用无人机巡检数据的获取电力设备缺陷信息,对巡检数据中存在的规律性、潜在性、趋势性问题和隐患进行自动判断识别,对数据进行关联分析和综合应用,可以深入地判断电力设备健康状态水平,并为运维部门提供更为准确的决策依据。

Power equipment has the characteristics of many points,wide area,long-term exposure to the field,etc.,and is easily affected by many factors such as the external environment during operation.External environ-mental factors include wind deflection,bird damage,icing,lightning strikes,wildfires,etc.Other factors in-clude equipment aging,quality,design,acceptance,operation and maintenance,etc.A single factor or multiple comprehensive factors can lead to equipment failure,and then cause large-scale power outages,seriously affect-ing equipment operation.With the application and development of UAV power equipment inspection,UAV in-spection data continues to accumulate.Through intelligent processing methods,UAV inspection data is used to obtain power equipment defect information.It can automatically judge and identify the regularity,potential,trend problems and hidden dangers of the data,and conduct correlation analysis and comprehensive application of the data,which can deeply judge the health status of power equipment and provide more accurate decision making basis for the operation and maintenance department.