期刊名称：Journal of Urban and Environmental Engineering
印刷版ISSN：1982-3932
电子版ISSN：1982-3932
出版年度：2017
卷号：11
期号：1
页码：119-130
语种：English
出版社：Universidade Federal da Paraíba
摘要：Successful modeling of hydro-environmental processes widely rel ies on quantity and quality of accessible data and noisy data might effect on the f unctioning of the modeling. On the other hand in training phase of any Artificial Intelligence (AI) based model, each training data set is usually a limited sample of po ssible patterns of the process and hence, might not show the behavior of whole populat ion. Accordingly in the present article first, wavelet-based denoising method was u sed in order to smooth hydrological time series and then small normally distributed no ises with the mean of zero and various standard deviati ons were generated and added t o the smoothed time series to form different denoised-jittered training data sets, for Artificial Neural Network (ANN) and Adaptive Neuro-Fuzzy Inference System (ANFIS) modeling of daily rainfall – runoff process of the Oconee River watershed l ocated in USA. To evaluate the modeling performance, the outcomes were compared w ith the results of multi linear regression (MLR) and Auto Regressive Integrated Mo ving Average (ARIMA) models. Comparing the achieved results via the trained ANN and ANFIS models using denoised-jittered data showed that the proposed da ta processing approach which serves both denoising and jittering techniques could impr ove performance of the ANN and ANFIS based rainfall-runoff modeling of the Oconee Rive r Watershed up to 13% and 11% in the verification phase.
其他摘要：Successful modeling of hydro-environmental processes widely rel ies on quantity and quality of accessible data and noisy data might effect on the f unctioning of the modeling. On the other hand in training phase of any Artificial Intelligence (AI) based model, each training data set is usually a limited sample of po ssible patterns of the process and hence, might not show the behavior of whole populat ion. Accordingly in the present article first, wavelet-based denoising method was u sed in order to smooth hydrological time series and then small normally distributed no ises with the mean of zero and various standard deviati ons were generated and added t o the smoothed time series to form different denoised-jittered training data sets, for Artificial Neural Network (ANN) and Adaptive Neuro-Fuzzy Inference System (ANFIS) modeling of daily rainfall – runoff process of the Oconee River watershed l ocated in USA. To evaluate the modeling performance, the outcomes were compared w ith the results of multi linear regression (MLR) and Auto Regressive Integrated Mo ving Average (ARIMA) models. Comparing the achieved results via the trained ANN and ANFIS models using denoised-jittered data showed that the proposed da ta processing approach which serves both denoising and jittering techniques could impr ove performance of the ANN and ANFIS based rainfall-runoff modeling of the Oconee Rive r Watershed up to 13% and 11% in the verification phase.
