A New Framework for Adptive Sampling and Analysis During Long-Term Monitoring and Remedial Action Management

Minsker, Barbara

University of Illinois at Urbana-Champaign

关键词: Calibration;    Management;    Remedial Action;    Genetics;    Monitoring;   

DOI  :  10.2172/894014
RP-ID  :  EMSP-87023B-2005
RP-ID  :  None
RP-ID  :  894014

美国|英语

来源: UNT Digital Library

【 摘 要 】

Yonas Demissie, a research assistant supported by the project, has successfully created artificial data and assimilated it into coupled Modflow and artificial neural network models. His initial findings show that the neural networks help correct errors in the Modflow models. Abhishek Singh has used test cases from the literature to show that performing model calibration with an interactive genetic algorithm results in significantly improved parameter values. Meghna Babbar, the third research assistant supported by the project, has found similar results when applying an interactive genetic algorithms to long-term monitoring design. She has also developed new types of interactive genetic algorithms that significantly improve performance. Gayathri Gopalakrishnan, the last research assistant who is partially supported by the project, has shown that sampling branches of phytoremediation trees is an accurate approach to estimating soil and groundwater contaminations in areas surrounding the trees at the Argonne 317/319 site.