基于Elman神经网络的传感器补偿算法研究  

Study of Sensor Compensation Algorithm Based on Elman Neural Network

作　　者：王秀芳[1] 句莉莉[1] 魏春明[2] 张延光[3] 

机构地区：[1]大庆石油学院电气信息工程学院,大庆163318 [2]天津大港油田定向井公司,天津300280 [3]中国石油冀东油田分公司供应处,唐海063200

出　　处：《科学技术与工程》2009年第20期5987-5990,共4页Science Technology and Engineering

基　　金：黑龙江省科技计划项目(GZ07A102)资助

摘　　要：为消除随钻测斜仪中的传感器由于受温度、湿度等非目标参量影响,提高随钻测斜仪的测量精度和工作稳定性,采用Elman神经网络,用LM算法对其进行训练,并将其应用到随钻测斜仪的传感器补偿中。仿真结果表明,精度可达10-7,比原来提高了4个数量级,提高了随钻测斜仪的测量精度和稳定性。该方法补偿速度快,补偿效果好,可以应用于其它各类传感器的补偿中。In order to eliminate the impact of sensor in LWD due to temperature, humidity and other non-target impact parameters, to improve the accuracy and stability of LED, the Elman neural network is used, using LM algorithm to train it, then its application is described in LWD inclinometer sensor compensation. Simulation results show that the iterative error accuracy is 10^-7, improved four orders of magnitude, improving the measurement accuracy and stability. The method of compensation is faster than other methods and has good effect and can be applied to other types of sensor compensation.

关 键 词：随钻测斜仪 传感器 补偿 Elman神经网络LM算法 

分 类 号：TE928[石油与天然气工程—石油机械设备]