| DOI: | 10.1016/j.marpolbul.2017.01.045
|
| Scopus记录号: | 2-s2.0-85011082484
|
| 论文题名: | Field measurements and neural network modeling of water quality parameters |
| 作者: | Haghiabi A.H.
|
| 刊名: | Marine Pollution Bulletin
|
| ISSN: | 0025-326X
|
| EISSN: | 1879-3363
|
| 出版年: | 2016
|
| 语种: | 英语
|
| 英文关键词: | Artificial neural networks
; Karkheh catchment
; Tireh River
; Water quality management
|
| Scopus关键词: | Catchments
; Forecasting
; Multilayer neural networks
; Neural networks
; Parameter estimation
; Quality management
; Water conservation
; Water management
; Water quality
; Water resources
; Water supply
; Field measurement
; Natural streams
; Neural network model
; Sampling stations
; Specific conductivity
; Total dissolved solids
; Water quality parameters
; Water quality predictions
; Rivers
; artificial neural network
; catchment
; conductance
; Iran
; nerve cell
; pH
; prediction
; river
; sampling
; sigmoid
; water quality
|
| Scopus学科分类: | Agricultural and Biological Sciences: Aquatic Science
; Earth and Planetary Sciences: Oceanography
; Environmental Science: Pollution
|
| 英文摘要: | Rivers are one of the main sources of water supply for agricultural, industrial, and urban use, and therefore, surveying their water quality is necessary. Recently, artificial neural networks have been proposed as a powerful tool for modeling and predicting water quality parameters in natural streams. In this paper, a multilayer neural network (MLP) model was developed to predict water quality parameters of Tireh River located in South West of Iran. The main parameters of water quality, namely total dissolved solids; specific conductivity; pH; and HCO3, Cl, Na, SO4, Mg, and Ca concentrations, were measured and predicted using the MLP model. The architecture of the proposed MLP model included two hidden layers, i.e., first and second hidden layers, in which eight and six neurons were considered, respectively. The tangent sigmoid and pure line functions were selected as transfer functions for the neurons in the hidden and output layers, respectively. Results showed that the MLP model performed suitably to predict water quality parameters of Tireh River. To assess the performance of the MLP model for water quality prediction along the studied area, in addition to the existing sampling stations, another 14 stations were considered. Evaluation of the performance of the developed MLP model to map the relationship between the water quality parameters along the studied area showed that the MLP model has suitable accuracy, and the minimum correlation between the results of the MLP model and the measured data was 0.85. © 2017. |
| Citation statistics: |
|
| 资源类型: | 期刊论文
|
| 标识符: | http://119.78.100.158/handle/2HF3EXSE/86863
|
| Appears in Collections: | 过去全球变化的重建
全球变化的国际研究计划
|
|
There are no files associated with this item.
|
|
作者单位: | Water Engineering Department, Lorestan University, Khorramabad, Iran
|
| Recommended Citation: | |
Haghiabi A.H.. Field measurements and neural network modeling of water quality parameters[J]. Marine Pollution Bulletin,2016-01-01
|
|
|