



摘"要:研究了基于改進YOLOv4的智能變電站二次設備穩(wěn)態(tài)AR紅外測溫技術,實現智能變電站二次設備穩(wěn)態(tài)識別。在分析紅外測溫原理的基礎上,利用AR紅外測溫眼鏡采集智能變電站二次設備溫度的紅外圖像,通過改進加權引導濾波增強算法對采集智能變電站二次設備溫度的紅外圖像實施圖像增強處理,將處理后的紅外圖像輸入改進YOLOv4網絡中,實現智能變電站二次設備穩(wěn)態(tài)識別。實驗表明:該方法可準確采集智能變電站二次設備8個保護屏柜的內部溫度,且接近設備實際運行溫度;同時有效降低智能變電站二次設備的紅外圖像的噪聲,對智能變電站二次設備穩(wěn)態(tài)識別的應用效果較好,能準確識別出智能變電站的二次設備異常狀態(tài),助力智能變電站運維管理員針對設備異常狀態(tài)做出對應維修方案。
關鍵詞:改進YOLOv4;智能變電站;二次設備穩(wěn)態(tài);AR紅外;測溫技術;損失函數
中圖分類號:TN271""""""文獻標識碼:A
Steady"State"AR"Infrared"Temperature"Measurement"
Technology"of"Secondary"Equipment"in"Intelligent"Substation"
Based"on"Improved"YOLOv4
LIU"Hao,MA"Qiang,FU"Qiang,CHEN"Yuan,ZHAO"Tonghan
(Guyuan"Power"Supply"Company"of"State"Grid"Ningxia"Electric"Power"Co.,"Ltd.,Guyuan,Ningxia"756000,China)
Abstract:The"steady"state"AR"infrared"temperature"measurement"technology"of"secondary"equipment"in"intelligent"substation"based"on"improved"YOLOv4"is"studied"to"realize"the"steady"state"identification"of"secondary"equipment"in"intelligent"substation."On"the"basis"of"analyzing"the"principle"of"infrared"temperature"measurement,"the"AR"infrared"temperature"measuring"glasses"are"used"to"collect"the"infrared"image"of"the"temperature"of"the"secondary"equipment"in"the"intelligent"substation,"and"the"infrared"image"of"the"temperature"of"the"secondary"equipment"in"the"intelligent"substation"is"enhanced"by"the"improved"weighted"guided"filter"enhancement"algorithm."The"processed"infrared"image"is"input"into"the"improvednbsp;YOLOv4"network"to"realize"the"steady"recognition"of"the"secondary"equipment"in"the"intelligent"substation."The"experiment"shows"that"this"method"can"accurately"collect"the"internal"temperature"of"8"protection"panels"of"the"secondary"equipment"in"the"intelligent"substation,"and"is"close"to"the"actual"operating"temperature"of"the"equipment;"At"the"same"time,"it"can"effectively"reduce"the"noise"of"the"infrared"image"of"the"secondary"equipment"of"the"intelligent"substation,"and"has"a"good"application"effect"on"the"steadystate"identification"of"the"secondary"equipment"of"the"intelligent"substation."It"can"accurately"identify"the"abnormal"status"of"the"secondary"equipment"of"the"intelligent"substation,"and"help"the"operation"and"maintenance"administrator"of"the"intelligent"substation"to"make"corresponding"maintenance"plans"for"the"abnormal"status"of"the"equipment.
Key"words:improved"YOLOv4;"intelligent"substation;"steady"state"of"secondary"equipment;"AR"infrared;"temperature"measurement"technology;"loss"function
智能變電站是國家電網的基礎設施,智能變電站設備穩(wěn)態(tài)運行對電網安全運行至關重要[1-3]。智能變電站二次設備是智能變電站系統(tǒng)的核心部分,主要包括保護裝置、通信裝置、智能終端等,對智能變電站一次設備起到保護的作用[4]。智能變電站二次設備的狀態(tài)識別作為智能變電站二次設備檢修的關鍵,迫切需要針對智能變電站二次設備狀態(tài)識別展開研究,這對電網安全運行具有深遠的意義[5,6]。但是以往智能變電站二次設備狀態(tài)識別方法非常局限,如葉遠波等[7]研究多模型融合集成學習的狀態(tài)評估方法,利用雙層基學習器實現對智能變電站二次設備狀態(tài)的評估。……