油田现场传感器在线校准与维护关键技术研究

Research on Key Technologies of Online Calibration and Maintenance of Oil Field Sensors

摘要: 针对国内大部分油田未开展计量数据准确性测试，甚至仍旧停留在计量仪表拆卸送检的传统方式即实验室送检和现场测试存在的问题。应用数理统计、神经网络、数据降维、可视化分析等技术对传感器历史数据进行数据挖掘，建立传感器失准失效诊断分析模型和失准定位模型，构建基于数据分析的仪表诊断分析平台；研发工业现场无线传感器数据非干扰接收解析技术，解决在用传感器数据获取问题；研发现场传感器在线校准技术和设备。提出的系统级数据分析、现场检定和在线校准三位一体的智能化综合解决方案，有利于减少传感器维护的工作量和技术难度，有利于保障工业信息化系统正常运行，提高运行效率。

Abstract: In view of the problems that the accuracy test of measurement data has not been carried out in most domestic oil fields, and even still stays in the traditional way of measuring instrument disassembly and inspection, that is, laboratory inspection and field test. We apply data mining of mathematical statistics, neural network, data dimension reduction and visual analysis to sensor historical data, establish sensor failure diagnosis analysis model and inaccurate positioning model, build the instrument diagnosis and analysis platform based on data analysis; research and develop the on-site wireless sensor data to solve the problem of using sensor data acquisition, and develop field sensor line calibration technology and equipment. The proposed intelligent comprehensive solution of system-level data analysis, field verification and online cali-bration is conducive to reduce the workload and technical difficulty of sensor maintenance, to en-sure the normal operation of industrial information system and improve the operation efficiency.