摘要
焊接质量直接影响天然气运输安全,为此研究了红外图像的长输天然气管道焊接裂纹检测系统,提升了裂纹检测效果。该系统利用超声波激发电路发送超声波至长输天然气管道,产生热量;通过红外图像采集卡采集由热量形成的红外图像,传输至计算机;计算机通过红外图像分析模块处理获取的信息,得到焊接裂纹检测结果,若属于焊接裂纹实施现场警报,在日志内存储焊接裂纹相关信息;红外图像分析模块利用热响应差分法增强红外图像,通过自适应区域生长方法分割增强的红外图像,得到异常温度区域,通过数学形态学提取原始红外图像的焊缝轮廓信息,中级融合异常温度区域与焊缝轮廓信息,提取裂纹。实验证明,该系统可精准检测管道焊接裂纹,裂纹提取的完整性与质量较高;在检测微小裂纹时,依旧具备较优的检测效果;该系统分割红外图像时的敏感性与特异度均较高,分割效果较佳。
Welding quality directly affects the safety of natural gas transportation.Therefore,the welding crack detection system of long distance natural gas pipeline with infrared image was studied to improve the crack detection effect.The system uses ultrasonic excitation circuit to send ultrasonic to long distance natural gas pipeline to generate heat.Infrared images formed by heat are collected by infrared image acquisition card and transmitted to PC.The PC processed the information obtained by infrared image analysis module,and obtained the welding crack detection results.If it belongs to the welding crack,the on-site alarm was implemented,and the welding crack related information was stored in the log.The infrared image analysis module uses the thermal response difference method to enhance the infrared image.The enhanced infrared image is segmented by the adaptive region growth method to obtain the abnormal temperature region.The weld contour information of the original infrared image is extracted by mathematical morphology,and the cracks are extracted by intermediate fusion of abnormal temperature region and weld contour information.Experiments showed that the system can accurately detect welding cracks in pipelines,and the integrity and quality of crack extraction are high.In detection of micro cracks,it still has a better detection effect.The system has high sensitivity and specificity in infrared image segmentation and good segmentation effect.
作者
叶朝伟
袁野
胡金宝
王新栩
Ye Chaowei;Yuan Ye;Hu Jinbao;Wang Xinxu(Zhejiang Zheneng Natural Gas Pipeline Network Co.,Ltd.,Hangzhou 310052,China)
出处
《能源与环保》
2022年第5期50-55,共6页
CHINA ENERGY AND ENVIRONMENTAL PROTECTION
基金
浙江省自然科学基金(GJJ191008)。
关键词
红外图像
长输天然气
管道焊接
裂纹检测
超声波
infrared image
long-distance natural gas transmission
pipeline welding
crack detection
ultrasonic wave
