摘要
铁路接触网使用复合绝缘子保障铁路电力正常供给已初有成效,但高原盐碱地区接触网绝缘子污闪事故时有发生。为辅助铁路供电部门对染污绝缘子及时清扫做好防污工作,通过试验得到不同型号的复合绝缘子在各种环境因素影响下的污闪电压,利用遗传算法优化BP神经网络建立接触网复合绝缘子污闪电压预测模型。将绝缘子型号因子(δ_(CF))、盐密(ESDD)、灰密(NSDD)、环境温度(T)、大气压强(P)五个特征量作为输入参数,污闪电压预测值为输出参数,经过仿真发现预测结果与试验结果之间偏差较小,绝缘子预测闪络电压十分接近试验闪络电压,仿真结果能够满足工程要求。该结果可帮助铁路部门及早发现供电安全隐患并处理,为高原盐碱地区接触网复合绝缘子防污闪工作积累宝贵经验。
The application of composite insulators to ensure the normal power supply of railway has achieved initial results,but the pollution flashover accidents of insulators in catenary network occur frequently in plateau salt and alkali areas.In order to assist the railway power supply department to clean the polluted composite insulators in time,the pollution flashover voltage of different types of composite insulators under the influence of various environmental factors is obtained through the test.The prediction model of pollution flashover voltage of composite insulators in catenary network is established by Optimizing BP neural network with genetic algorithm.Five characteristic parameters of insulators,i.e.Model parameters of insulators(δ_(CF)),Equivalent salt deposit density(ESDD),Non soluble deposit density(NSDD),Temperature(T),Atmospheric pressure(P),are taken as input parameters.The predicted value of insulators pollution flashover voltage is the output parameter.The simulation results show that the error between the prediction results and the test results is small,the predicted flashover voltage of insulator is very close to the test flashover voltage,and the simulation results can meet the engineering requirements.The results can help the railway department to find and deal with the hidden risk of power supply as early as possible,and accumulate valuable experience for the prevention of composite insulators pollution flashover in plateau salt and alkali area.
作者
王思华
王军军
赵珊鹏
WANG Sihua;WANG Junjun;ZHAO Shanpeng(School of Automation&Electrical Engineering,Lanzhou Jiaotong University,Lanzhou 730070,China;Rail Transit Electrical Automation Engineering Laboratory of Gansu Province,Lanzhou Jiaotong University,Lanzhou 730070,China)
出处
《电瓷避雷器》
CAS
北大核心
2022年第1期134-142,150,共10页
Insulators and Surge Arresters
基金
国家自然科学基金资助项目(编号:51767014,51867013)
中国铁路总公司科技研究开发计划资助项目(编号:2017010-C)。
关键词
接触网
复合绝缘子
电压预测
BP神经网络
遗传算法
高原盐碱区
catenary
composite insulator
voltage prediction
back propagation neural network
genetic algorithm
plateau saline alkali area
