摘要
由于井下工作环境的特殊性,井下电气设备出现接触不良等问题时,矿井供电系统的电气连接点处将产生串联故障电弧,引起严重的电气火灾。针对这一情况,设计了一种基于BP神经网络的井下电弧火灾预警方法。BP神经网络具有较强的系统适应性和学习能力,可精确捕捉井下串联型故障电弧的动态特性,在矿井供电系统出现潜伏的串联电弧时进行报警,从而避免了电气火灾事故的发生。结果表明:该方法可以精确地对井下串联故障电弧进行识别预警,减少了井下电气火灾事故的发生概率,为新型智能矿井工作环境的设计提供参考。
Due to the particularity of underground working environment, when there are problems such as poor contact of underground electrical equipment, series fault arcs will occur at the electrical connection points of mine power supply system, causing serious electrical fire.In view of this situation, a downhole arc fire warning method based on BP neural network is designed. With strong system adaptability and learning ability, BP neural network can accurately capture the dynamic characteristics of downhole series fault arcs, and give an alarm when latent series arcs appear in the mine power supply system, thus avoiding the occurrence of electrical fire accidents. The results show that this method can accurately identify and warn downhole series fault arc, reduce the probability of downhole electrical fire accident, and provide reference for the design of new intelligent mine working environment.
作者
葛明臣
刘大同
GE Ming-chen;LIU Da-tong(Hegang Mining Company,Longmay Mining Group Co.,Ltd.,Hegang 154100,China;School of Electrical and Control Engineering,Heilongjiang University of Science and Technology,Harbin 150022,China)
出处
《煤炭技术》
CAS
2020年第9期195-198,共4页
Coal Technology
关键词
井下电弧
电气火灾
BP神经网络
underground electric arc
electrical fire
BP neural network
