摘要
为实现油浸式变压器风扇运行状态异常的在线监测,提出了一种以自冷油循环变压器油指数为特征值的风扇风量异常监测方法。首先,为找出变压器在风扇运行状态变化前后的特征量区别,基于动量、能量、油流量守恒公式建立变压器油浸自冷(oil natural air natural,ONAN)和油浸风冷(oil natural air forced,ONAF)两种冷却方式的温升计算模型;然后,根据计算模型获得变压器不同负载的顶层油温升,利用油温拟合并对比两种冷却方式下的油指数,得到ONAN和ONAF冷却模式下油指数分别为0.8213和0.9280,说明油指数区别风扇风量异常前后状态的显著性;最后,基于粒子群算法(particle swarm optimization,PSO)对变压器顶层油温现场数据进行油指数在线监测。研究结果表明:风扇风量的减少使油指数从0.9~0.95区间过渡到0.8~0.85区间,能够较为灵敏的反映变压器风扇风量变化。该研究为风扇早期故障的智能化监测提供了新的思路和方法。
In order to realize online monitoring of abnormal fan operation state of oil-immersed transformers,a fan airflow anomaly monitoring method with the characteristic value of oil exponent of self-cooled oil circulation transformers is proposed.Firstly,in order to find out the difference of characteristic quantity of transformer before and after the change of fan operation state,a calculation model of temperature rise of two cooling modes for transformer,namely,oil natural air natural(ONAN)cooling and oil natural air forced(ONAF)cooling,is established based on the equation of momentum,energy and oil flow conservation.Then,based on the calculation model,the oil temperature rise in the top layer of the transformer with different loads is obtained,and the oil exponents of the two cooling modes are compared.It is obtained that the oil exponent is 0.8213 and 0.9280 for ONAN and ONAF cooling modes,respectively,which illustrates the significance of the oil exponent to distinguish the state before and after the abnormal fan airflow.Finally,the oil exponent is monitored online based on particle swarm optimization(PSO)for the field data of transformer top oil temperature.The research results show that the reduction of fan airflow enables the oil exponent to transit from 0.9~0.95 range to 0.8~0.85 range,which can reflect the change of transformer fan airflow more sensitively.The study provides a new idea and method for the intelligent monitoring of the early fan failure.
作者
左弯弯
张建文
王路伽
王冬伟
陈婷
题恒
ZUO Wanwan;ZHANG Jianwen;WANG Lujia;WANG Dongwei;CHEN Ting;TI Heng(School of Electrical and Power Engineering,China University of Mining and Technology,Xuzhou 221116,China)
出处
《高电压技术》
EI
CAS
CSCD
北大核心
2022年第7期2747-2753,共7页
High Voltage Engineering
基金
中央高校基本科研业务费专项资金(2021QN1064)
广东省基础与应用基础研究基金(2021A1515110435)。
关键词
油浸式变压器
温升热模型
油指数
状态监测
粒子群算法
oil-immersed transformer
temperature rise thermal model
oil exponent
condition monitoring
particle swarm algorithm
