波谱学杂志 ›› 2018, Vol. 35 ›› Issue (2): 226-233.doi: 10.11938/cjmr20172603

• 低场磁共振技术与应用专栏 • 上一篇    下一篇

基于蒙特卡罗算法的低场核磁共振测量效率提高方法

邹越崎1, 郭盼2, 徐征1   

  1. 1. 重庆大学, 输配电装备及系统安全与新技术国家重点实验室, 重庆 400044;
    2. 重庆师范大学 物理与电子工程学院, 重庆 401331
  • 收稿日期:2017-11-08 出版日期:2018-06-05 发布日期:2017-12-19
  • 通讯作者: 郭盼,Tel:13527475882,E-mail:guopan0822@163.com E-mail:guopan0822@163.com
  • 基金资助:
    国家自然科学基金资助项目(51677008,51377182,11647098,51707028).

High-Efficiency Low-Field Nuclear Magnetic Resonance Measurements with a Monte Carlo Simulation Algorithm

ZOU Yue-qi1, GUO Pan2, XU Zheng1   

  1. 1. State Key Laboratory of Power Transmission Equipment & System Security and New Technology, Chongqing University, Chongqing 400044, China;
    2. School of Physics and Electronic Engineering, Chongqing Normal University, Chongqing 401331, China
  • Received:2017-11-08 Online:2018-06-05 Published:2017-12-19

摘要: 核磁共振(NMR)的纵向弛豫时间(T1)、横向弛豫时间(T2)、自扩散系数(D0),以及T2-T1T2-D0测量目前广泛应用于石油测井行业.在测量D0的SGSE序列中,通过逐渐增大90°和180°脉冲之间的时间间隔(Td),可以对液体扩散行为产生的影响进行调节.然而Td的"起点"、"步进数"和"终点"等参数必须设置得当才能准确测量T1D0.目前参数的设置依赖多次的人工调整和测量人员的经验,耗时且使用门槛较高.本文用蒙特卡罗方法进行大量随机模拟,根据前面若干点的测量结果筛选出满足要求的随机值,预测下一个测量点的位置.该算法可以实时更新参数设置,实现自动化测量,达到降低测量门槛、缩短测量时间的目的.经验证,该算法可以适用于T1D0的测量.

关键词: 低场核磁共振, 算法, 蒙特卡罗, T1, D0

Abstract: Measurements of longitudinal relaxation time (T1), transverse relaxation time (T2), self-diffusion coefficient (D0),T2-T1 and T2-D0 are central in NMR-based oil logging. The SGSE sequence is commonly used to measure D0, in which the interval time between the 90° and 180° pulses (Td) is increased incrementally to probe into the diffusion behaviors of liquids. However, the starting point, step size and end point of Td must be properly set in order to get accurate measurement of T1 and D0. Currently, tuning of such parameters is often done manually, and thus time-consuming and difficult to use. The final outcome also relies heavily on operator's experience. In this study, a large number of random simulations were carried out by a Monte Carlo algorithm. The algorithm predicted the parameters for next measurement based on the results from previous measurements and thus was capable of updating the parameter settings in real time for automatic measurements. The algorithm was validated for T1, D0 measurements, and demonstrated reduced measurement threshold and shortened measurement time.

Key words: low-field NMR, algorithm, Monte Carlo, T1, D0

中图分类号: