摘要
针对传统抽水蓄能电站廊道渗水检测依旧采用人工检测方式,导致渗水检测识别准确率低、检测效率低,渗水检测结果不佳的问题,提出一种基于红外成像的抽水蓄能电站廊道渗水检测识别技术。首先,从硬件部分和软件部分搭建一个抽水蓄能电站廊道渗水检测设备,然后采用红外成像法对蓄能电站廊道渗水口进行特征提取和无损检测;最后利用该设备对各渗水点数据进行统计分析,从而对该廊道渗水安全进行评估。实验结果表明,提出的渗水检测识别方法的最大误差值和最大相对误差值分别为544 mm^(2)和1.51%,渗水检测误差较小,检测识别精度可达98.49%,识别精度较高,可在蓄能电站廊道渗水中进行广泛应用。
In view of the problems of water seepage detection of traditional pumped storage power station corridor still using manual detection method, which leads to low water seepage detection and identification accuracy, slow detection efficiency and poor water seepage detection results, a water seepage detection and identification technology of pumped storage power station corridor based on infrared imaging is proposed. First, a water seepage detection equipment of the pumped storage power station corridor is constructed from the hardware part and the software part, and then the infrared imaging method is used to extract features and NDT on the water seepage outlet of the energy storage power station. Finally, the data of the water seepage points are analyzed to evaluate the water seepage safety of the corridor. The experimental results show that the maximum error value and the maximum relative error value of the proposed water seepage detection and identification method are 544 mm^(2) and 1.51%, respectively, and the water seepage detection accuracy is 98.49% and high, which can be widely used in the water seepage of the energy storage power station corridor.
作者
孔繁臣
KONG Fanchen(Inner Mongolia Chifeng Pumped Storage Co.,LTD.,Chifeng 024000,China)
出处
《自动化与仪器仪表》
2022年第12期170-175,共6页
Automation & Instrumentation
基金
青年科学基金项目《抽水蓄能电站钢岔管月牙形几何边界周期性高应力的演化机制》(51609020)。
关键词
红外成像
抽水蓄能电站
渗水点
特征提取
检测识别
infrared imaging
pumped storage power station
water seepage point
feature extraction
detection and identification
