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

Classification of stormwater and lan...

Ha, Haejin.

FindBook

Google Book

Amazon

博客來

Classification of stormwater and landuse using neural networks.

紀錄類型:	書目-語言資料,印刷品 : Monograph/item
正題名/作者:	Classification of stormwater and landuse using neural networks./
作者:	Ha, Haejin.
面頁冊數:	123 p.
附註:	Chair: Michael K. Stenstrom.
Contained By:	Dissertation Abstracts International63-11B.
標題:	Engineering, Civil. -
電子資源:	http://pqdd.sinica.edu.tw/twdaoapp/servlet/advanced?query=3070094
ISBN:	0493897941

Classification of stormwater and landuse using neural networks.
Ha, Haejin.

Classification of stormwater and landuse using neural networks. - 123 p.

Chair: Michael K. Stenstrom.

Thesis (Ph.D.)--University of California, Los Angeles, 2002.

Stormwater runoff is a major contributor to the pollution of coastal waters in the United States. Differences in landuse patterns result in different pollutant concentrations, and therefore landuse-related control strategies are essential to control storm water pollution effectively. An approach that can differentiate landuse types in stormwater could provide opportunities for better landuse management to minimize stormwater pollution.

ISBN: 0493897941Subjects--Topical Terms:

783781
Engineering, Civil.

Classification of stormwater and landuse using neural networks.
LDR:03685nam 2200325 a 45 001 935888
005 20110510
008 110510s2002 eng d
020 $a 0493897941
035 $a (UnM)AAI3070094
035 $a AAI3070094
040 $a UnM $c UnM
100 1 $a Ha, Haejin. $3 1259590
245 1 0 $a Classification of stormwater and landuse using neural networks.
300 $a 123 p.
500 $a Chair: Michael K. Stenstrom.
500 $a Source: Dissertation Abstracts International, Volume: 63-11, Section: B, page: 5436.
502 $a Thesis (Ph.D.)--University of California, Los Angeles, 2002.
520 $a Stormwater runoff is a major contributor to the pollution of coastal waters in the United States. Differences in landuse patterns result in different pollutant concentrations, and therefore landuse-related control strategies are essential to control storm water pollution effectively. An approach that can differentiate landuse types in stormwater could provide opportunities for better landuse management to minimize stormwater pollution.
520 $a A neural network model was applied to examine the relationship between stormwater water quality and various types of landuse. The neural model can be used to identify landuse types for future known and unknown cases. The neural model uses a Bayesian network and has ten water quality input variables, four neurons in the hidden layer, and five landuse target variables. The neural model correctly classified 92.3% of test files. Simulations were performed to predict the landuse type of a known data set, and accurately described the behavior of the new data set. This study demonstrates that a neural network can effectively classify landuse types with water quality data.
520 $a A similar approach was applied to two local stormwater monitoring programs, which use human activity as measured by landuse or standard industrial classification code to describe stormwater, in the hopes that these classifications will be useful to planners and regulators in abating stormwater pollution. Data sets produced by the landuse based program were successfully identified by the neural network, and the monitoring program is successful in accomplishing its goals. The industrial stormwater monitoring program is not successful; standard industrial classification code is not related to stormwater quality. Improvements are suggested, which include sample type, parameters and timing.
520 $a Beach water quality monitoring programs were next evaluated using examples from Southern California. Data collection that is useful in minimizing fecal contamination from human and animal waste is their major objective. Lack of specificity of indicator organisms is the major problem, and better, real-time indicators are needed. Beach monitoring programs seldom collect water quality or other data that might be used with neural network techniques to identify pollutant sources. Methods to detect human fecal pollution and differentiate it from other sources such as animals are reviewed. Microbial methods, especially those using molecular biology, and chemical methods are reviewed. At present there is no easy, low cost or rapid method for differentiating between human and non-human fecal contamination. It is much more likely that a combination of methods can be used to accurately identify human fecal pollution.
590 $a School code: 0031.
650 4 $a Engineering, Civil. $3 783781
650 4 $a Engineering, Environmental. $3 783782
650 4 $a Engineering, Sanitary and Municipal. $3 1018731
690 $a 0543
690 $a 0554
690 $a 0775
710 2 0 $a University of California, Los Angeles. $3 626622
773 0 $t Dissertation Abstracts International $g 63-11B.
790 $a 0031
790 1 0 $a Stenstrom, Michael K., $e advisor
791 $a Ph.D.
792 $a 2002
856 4 0 $u http://pqdd.sinica.edu.tw/twdaoapp/servlet/advanced?query=3070094