基于油浸式机器人的变压器故障识别研究

Research on transformer fault identification based on oil-immersed robot

油浸式变压器是电力系统中最为关键和核心的设备,定期检修的主要方式是停机放油后工人进入变压器内部目视检查,存在检修效率不高、停机时间长、故障识别准确性低等问题。本文研究采用新型智能机器人技术设计电力变压器油浸式智能内检机器人,在大型电力变压器不放油的情况下,进入停运设备内部主绝缘间隙仿生潜游,利用配备的摄像机和传感器将变压器内部视频图像等信息实时无线传输到工作站,使用卷积神经网络对内部的异物、闪络、放电点及痕迹进行智能感知和自动识别。试验结果表明,采用油浸式智能机器人的故障智能识别更准确、使用更方便,极大提高了电力变压器检修领域的智能化水平。

The oil-immersed transformer is the most critical and core equipment in the power system.The main way of regular maintenance is to enter the transformer for visual inspection after the oil is drained after shutdown.There are problems such as low maintenance efficiency,long shutdown time,and low accuracy of fault identification.In this paper,a new type of intelligent robot technology is used to design an oil-immersed intelligent internal inspection robot for power transformers.Under the condition that large power transformers do not drain oil,the robot can enter the main insulation gap of the out-of-service equipment for bionic diving,and use the equipped cameras and sensors to monitor the inside of the transformer.Video images and other information are wirelessly transmitted to the workstation in real time,and the convolutional neural network is used to perform intelligent perception and automatic identification of internal foreign objects,flashovers,discharge points and traces.The test results show that the intelligent identification of faults using the oil-immersed intelligent robot is more accurate and more convenient,which greatly improves the intelligence level in the field of power transformer maintenance.