摘要
原油含水率的测量对于确定油井出水出油层位、掌握油田地层动态、预计寿命等具有十分重要的意义。基于微波相移法的原油含水率检测技术,因其在检测过程中不受油水状态影响的显著优势,而受到越来越多的关注。原油中的矿化度伴生气和环境温度的变化会影响传感器的检测精度。因此,建立矿化度和温度的误差校正模型至关重要。对不同含量的氯化钾在不同温度下对微波法含水仪检测误差的影响规律进行了研究。在试验测量的数据的基础上,运用径向基(RBF)神经网络建立误差校正模型,实现误差的实时校正。测试结果表明,检测误差由原来的±25.2%降到±4.98%以内。由此可知,使用RBF神经网络可减少矿化度和温度对传感器检测精度的影响,在工程实践具中有重要的参考意义。
The detection of the water-cut of crude oil is of great significance for determining the water layer and oil layer of oil well outlet,understanding the formation dynamics of oil field and predicting the lifetime of oil field.Microwave-phase-shift technique for moisture content detection of crude oil has attracted more attention,because it is not affected by the state of oil-water during detection process.The variation of salinity,associated gas content and ambient temperature in crude oil will affect the detection accuracy of the microwave water content sensor.Therefore,it is very important to establish the error correction model of salinity and temperature.The influence of different content of potassium chloride at different temperature on the measurement error of microwave water-cut meter is accounted.Based on the experimental data,the RBF neural network is used to establish the real-time error correction model.The results showed that the detection error decreased from±25.2%to±4.98%.The use of radial basis function(RBF)neural network can reduce the influence of salinity and temperature on the detection accuracy of the sensor,and has great reference significance in engineering practice.
作者
董鹏敏
郭铅铅
王天琦
赵海空
罗仕冲
曾祥虎
王鹏
DONG Pengmin;GUO Qianqian;WANG Tianqi;ZHAO Haikong;LUO Shichong;ZENG Xianghu;WANG Peng(School of Mechanical Engineering,Xi'an Shiyou University,Xi'an 710065,China)
出处
《自动化仪表》
CAS
2020年第12期23-26,33,共5页
Process Automation Instrumentation
基金
陕西省自然科学基础研究计划重大基础研究基金资助项目(2016ZDJC-11)。
关键词
原油含水率
相移法
微波法含水仪
检测精度
温度
氯化钾
径向基函数神经网络
误差校正
Water-cut of crude oil
Phase shift method
Microwave water content sensor
Detection accuracy
Temperature
KCL
Radial basis fuction(RBF)neural network
Error correction
