科研之友
微信
新浪微博
Facebook
分享链接
摘要
Monitoring the extent and evolution of contaminant plumes in local and regional groundwater systems from existing landfills is critical in contamination control and remediation. The self-potential survey is an efficient and economical nondestructive geophysical technique that can be used to investigate underground contaminant plumes. Based on the unscented transform, we have built a Kalman filtering cycle to conduct time-lapse data assimilation for monitoring the transport of solute. based on the solute transport experiment using a bench- scale physical model. The data assimilation was formed by modeling the evolution based on the random walk model and observation correcting based on the self-potential forward. Thus,. monitoring self-potential data can be inverted by the data assimilation technique. As a result, we can reconstruct the dynamic process of the contaminant plume instead of using traditional frame-to-frame static inversion, which. may cause inversion artifacts. The data assimilation inversion algorithm was evaluated through noise-added synthetic time-lapse self-potential data. The result of the. numerical experiment shows validity, accuracy. and tolerance to the noise of the dynamic inversion. To validate the proposed algorithm, we conducted a scaled-down sandbox self-potential observation experiment to generate time-lapse data that closely mimics the. real-world contaminant monitoring setup. The results of physical experiments support. the idea. that the data assimilation method is a potentially useful approach for characterizing the transport of contamination plumes. using the unscented Kalman filter (UKF) data assimilation technique applied to. field time-lapse self-potential data.
- 出版日期2017-8
- 单位中南大学
