東華大學圖書館 |

語系: 繁體中文

說明(常見問題)

回圖書館首頁

手機版館藏查詢

登入

回首頁

切換: 標籤 | MARC模式 | ISBD

Using stochastic and genetic method ...

He, Changming.

FindBook

Google Book

Amazon

博客來

Using stochastic and genetic method to study prediction uncertainty for solute transport in heterogeneous media: Evaluation, conditioning and optimization.

紀錄類型:	書目-語言資料,印刷品 : Monograph/item
正題名/作者:	Using stochastic and genetic method to study prediction uncertainty for solute transport in heterogeneous media: Evaluation, conditioning and optimization./
作者:	He, Changming.
面頁冊數:	208 p.
附註:	Adviser: Bill X. Hu.
Contained By:	Dissertation Abstracts International67-03B.
標題:	Hydrology. -
電子資源:	http://pqdd.sinica.edu.tw/twdaoapp/servlet/advanced?query=3210922
ISBN:	9780542602832

Using stochastic and genetic method to study prediction uncertainty for solute transport in heterogeneous media: Evaluation, conditioning and optimization.
He, Changming.

Using stochastic and genetic method to study prediction uncertainty for solute transport in heterogeneous media: Evaluation, conditioning and optimization. - 208 p.

Adviser: Bill X. Hu.

Thesis (Ph.D.)--University of Nevada, Reno, 2006.

Considering the calculation efficiency, the method of moments apparently has a superior speed.

ISBN: 9780542602832Subjects--Topical Terms:

545716
Hydrology.

Using stochastic and genetic method to study prediction uncertainty for solute transport in heterogeneous media: Evaluation, conditioning and optimization.
LDR:03889nam 2200313 a 45 001 971852
005 20110927
008 110927s2006 eng d
020 $a 9780542602832
035 $a (UnM)AAI3210922
035 $a AAI3210922
040 $a UnM $c UnM
100 1 $a He, Changming. $3 1295881
245 1 0 $a Using stochastic and genetic method to study prediction uncertainty for solute transport in heterogeneous media: Evaluation, conditioning and optimization.
300 $a 208 p.
500 $a Adviser: Bill X. Hu.
500 $a Source: Dissertation Abstracts International, Volume: 67-03, Section: B, page: 1342.
502 $a Thesis (Ph.D.)--University of Nevada, Reno, 2006.
520 $a Considering the calculation efficiency, the method of moments apparently has a superior speed.
520 $a Accurate prediction of groundwater flow and contaminant transport depends on accurate understanding of aquifer's properties, such as hydraulic conductivity distribution. However, due to technique or economic reasons, a perfect known conductivity distribution is very hard, if not impossible, to obtain in practice. This stimulates the quick growth of stochastic hydrology and parameters calibration techniques.
520 $a In this dissertation, both these two areas are involved. In the first part, a numerical moment method (NMM) is extended and applied to study solute transport in a multi-scale formation with physical and chemical heterogeneities. In this method, a general expression of covariance function of hydraulic conductivity or sorption coefficient is derived for a multi-scale composite heterogeneous formation, based on the statistical properties of composite materials. Then the statistics of solute flux (mean and variance) that pass through a control plan are derived and solved by a numerical moments method. Several cases with different synthetic formations are studied and compared with Monte Carlo method. The results show that the mean value of total solute flux calculated by the two methods matched each other very well, but the variance of total solute flux obtained by the method of moments is smaller than that by the Monte Carlo method, especially for the cases with large total variances of the conductivity and sorption coefficient.
520 $a In the second part, we focus on the hydraulic conductivity identification and optimal tracer test design. Firstly, a gradient-based inverse technique, sequential self-calibration (SSC) method is introduced to map the hydraulic conductivity distribution conditioning to tracer test data. This method uses a streamline transport simulator to simulate tracer movement and combines a kriging approach to take the consideration of geostatistic properties of concerned parameters. Several cases were studied on a synthetic field by applying SSC method, and the results prove it an accurate and efficient approach to estimate the conductivity distribution.
520 $a The extended SSC method was then used as a basis to investigate the optimization of tracer test design. Two design factors are considered in this study separately: sampling time and observation well locations. We define a scaled information matrix as our optimality criterion, which is equivalent to a sum of squared scaled-sensitivities, employing the observation that parameters are most accurately estimated at points with high sensitivity to the parameters. To maximize the optimality criterion, a genetic algorithm is applied to search the optimal tracer test design. And finally, the influence of a sandwich-like geologic structure on the optimal observation well locations is also studied.
590 $a School code: 0139.
650 4 $a Hydrology. $3 545716
690 $a 0388
710 2 0 $a University of Nevada, Reno. $3 626634
773 0 $t Dissertation Abstracts International $g 67-03B.
790 $a 0139
790 1 0 $a Hu, Bill X., $e advisor
791 $a Ph.D.
792 $a 2006
856 4 0 $u http://pqdd.sinica.edu.tw/twdaoapp/servlet/advanced?query=3210922