摘要
汇控柜内温度过高可能导致柜内设备出现装置故障、绝缘损坏等问题。为实现对汇控柜内整体温度的精准把控和局部发热异常的快速消除,基于卷积神经网络研制了汇控柜智能降温装置,通过红外传感器扫描获得汇控柜内温度图像数据,控制芯片基于卷积神经网络模型分析定位发热位置,并控制散热模块对发热部位进行有效降温。现场试验结果表明,本装置可有效降低汇控内平均温度值和最高温度值,提高了智能设备的运行可靠性。
Excessive temperature inside the control cabinet tend to cause device faults,insulation damage,etc.In order to achieve accurate control of the overall temperature in the control cabinet and the quick elimination of local heating anomalies,the intelligent cooling device for the control cabinet is developed based on convolutional neural network.The temperature image data in the control cabinet is obtained by infrared sensor scanning;then,the control chip analyzes and locates the heating position based on the convolutional neural network model,and controls the cooling module to effectively cool the heating part.The field test results show that the device can effectively reduce the average temperature and the maximum temperature in the control cabinet,thus improving the operation reliability of the intelligent equipment.
作者
常俊晓
应宇鹏
胡方恺
马秀林
CHANG Junxiao;YING Yupeng;HU Fangkai;MA Xiulin(Taizhou Power Supply Company,State Grid Zhejiang Electric Power Co.,Ltd.,Taizhou Zhejiang 318000,China)
出处
《湖北电力》
2023年第3期76-81,共6页
Hubei Electric Power
基金
国网浙江省电力有限公司群众性创新项目(项目编号:5211TZ21000V)。
关键词
汇控柜
降温装置
红外传感器
卷积神经网络
控制芯片
control cabinet
cooling device
infrared sensor
convolutional neural network
control chip
