基于神经网络的配电柜电气火灾预警模型设计  

Design of Electric Fire Warning Model for Distribution Cabinet Based on Neural Network

作　　者：贾振国[1] 许琳[2] 杨忠亮 

机构地区：[1]长春工程学院能源动力工程学院,吉林长春130012 [2]长春工程学院计算机技术与工程学院,吉林长春130012 [3]南方电网公司深圳市供电局有限公司,广东深圳518101

出　　处：《科技创新与应用》2018年第10期91-93,共3页Technology Innovation and Application

基　　金：吉林省科技发展计划"民用新能源发电并网检测控制装置关键技术研究"(编号:20150204044SF);吉林省科技发展计划"电气火灾预警系统关键技术研究及应用"(编号:20160204019SF);住房和城乡建设部科学技术项目计划"民用新能源发电并网检测控制装置研发"(编号:2014-K1-067);住房和城乡建设部科学技术项目计划"建筑配电电能质量监测及漏电火灾预警系统研发"(编号:2014-K8-064)

摘　　要：配电柜是电力系统及工矿企业重要的配电设备之一,其安全、稳定运行直接关系到企业的正常生产和人们的日常生活。配电柜电气火灾是配电柜运行过程中的重要安全隐患。针对配电柜电气火灾的成因,提出了基于神经网络的配电柜电气火灾预警模型。该模型在考虑常规电气火灾诱发因素如弧光、烟雾、温度等的同时,融合了电压、电流和电能质量等电气参量作为电气火灾成因的特征量。该模型具有自学习、记忆和推理判断的特点,能够有效提高预警的可靠性。采用MATLAB仿真验证了模型预警的效果。Distribution cabinet is one of the important distribution equipment in power system and industrial and mining enterprises.Its safe and stable operation is directly related to the normal production of enterprises and people's daily life.Electrical fire of distribution cabinet is an important safety hazard in the operation of distribution cabinet.In view of the causes of electrical fire in distribution cabinets,an electric fire warning model of distribution cabinets based on neural network is proposed.In this model,the electrical parameters,such as voltage,current and power quality,are considered as the characteristics of electrical fire caused by conventional electrical fire inducing factors,such as arc,smoke,temperature and so on.The model has the characteristics of selflearning,memory and reasoning and judgment,and can effectively improve the reliability of early warning.The effect of model warning is verified by MATLAB simulation.

关 键 词：配电柜 神经网络 电气火灾 预警 模型 

分 类 号：TP216.3[自动化与计算机技术—检测技术与自动化装置]