Multi-step water quality forecasting using a boosting ensemble multi-wavelet extreme learning machine model

Barzegar, Rahim; Moghaddam, Asghar Asghari; Adamowski, Jan; Ozga-Zieliński, Bogdan

doi:10.1007/s00477-017-1394-z

Cited by 100 publications

(43 citation statements)

References 51 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…This study showed a greater overall performance of hybrid wavelet transformed ELM (WA-ELM) model as compared to the ELM model. Using a boosting ensemble multi-wavelet extreme learning machine (Multi-WA-ELM) model improved water quality forecasting as compared to individual WA-ELM and ELM models [BARZEGAR et al 2017].…”

Section: Introductionmentioning

confidence: 99%

Application of SARIMA model to forecasting monthly flows in Waterval River, South Africa

Tadesse

Dinka

2017

Journal of Water and Land Development

View full text Add to dashboard Cite

Knowledge of future river flow information is fundamental for development and management of a river system. In this study, Waterval River flow was forecasted by SARIMA model using GRETL statistical software. Mean monthly flows from 1960 to 2016 were used for modelling and forecasting. Different unit root and Mann-Kendall trend analysis proved the stationarity of the observed flow time series. Based on seasonally differenced correlogram characteristics, different SARIMA models were evaluated; their parameters were optimized, and diagnostic check up of forecasts was made using white noise and heteroscedasticity tests. Finally, based on minimum Akaike Information (AI) and Hannan-Quinn (HQ) criteria, SARIMA (3, 0, 2) x (3, 1, 3) 12 model was selected for Waterval River flow forecasting. Comparison of forecast performance of SARIMA models with that of computational intelligent forecasting techniques was recommended for future study.

show abstract

Section: Introductionmentioning

confidence: 99%

Application of SARIMA model to forecasting monthly flows in Waterval River, South Africa

Tadesse

Dinka

2017

Journal of Water and Land Development

View full text Add to dashboard Cite

show abstract

“…In fact, the Lunj-Box test for the non-correlation hypothesis does not reject, using up to 30 different lags in all the locations under study, and Figure 4 presents the histograms of the residuals that resemble the normal curve. 9.599 0.00 0.00 9.983 0.00 0.00 9.673 0.00 0.00 9.559 0.00 0.00 9.532 0.00 0.00 β 3 9.212 0.00 0.00 9.546 0.00 0.00 9.138 0.00 0.00 9.504 0.00 0.00 9.737 0.00 0.00 β 4 8.753 0.00 0.00 9.001 0.00 0.00 9.356 0.00 0.00 9.134 0.00 0.00 9.014 0.00 0.00 β 5 8.313 0.00 0.00 9.262 0.00 0.00 8.555 0.00 0.00 8.688 0.00 0.00 8.753 0.00 0.00 β 6 8.362 0.00 0.00 7.771 0.00 0.00 7.270 0.00 0.00 7.847 0.00 0.00 7.939 0.00 0.00 β 7 7.273 0.00 0.00 8.164 0.00 0.00 7.271 0.00 0.00 8.576 0.00 0.00 8.376 0.00 0.00 β 8 7.087 0.00 0.00 8.844 0.00 0.00 6.556 0.00 0.00 7.668 0.00 0.00 8.260 0.00 0.00 β 9 7.971 0.00 0.00 7.941 0.00 0.00 5.800 0.00 0.00 7.166 0.00 0.00 7.319 0.00 0.00 β 10 7.410 0.00 0.00 7.844 0.00 0.00 7.673 0.00 0.00 7.561 0.00 0.00 8.477 0.00 0.00 β 11 8.030 0.00 0.00 8.968 0.00 0.00 9.737 0.00 0.00 9.673 0.00 0.00 9.408 0.00 0.00 β 12 9.859 0.00 0.00 9.946 0.00 0.00 9.418 0.00 0.00 9.025 0.00 0.00 9.207 0.00 0.00 Furthermore, with the exception of the CAR location, the residuals of the calibration model do not reject (at a 1% significance level) the normality assumption using the Jarque-Bera test or the Kolmogorov-Smirnov test; the K-S p-values are presented in Table 5.…”

Section: Resultsmentioning

confidence: 99%

“…In [6], generalized additive models of location, scale, and shape (GAMLSS) were applied to characterize model uncertainty, due to incomplete understanding of physical processes, in an Atlantic coastal plain watershed system. In [7], extreme learning machine (ELM) and wavelet-extreme learning machine hybrid (WA-ELM) models were applied to forecast multi-step-ahead electrical conductivity (EC)-a water quality indicator that is useful for estimating the mineralization and salinity of water-and to employ an integrated method to combine the advantages of WA-ELM models, which utilized the boosting ensemble method. Control charts were developed by [8] to treat the case of a French river, for which the parameter of interest, the dissolved oxygen concentration (DO), was characterized by a non-stationary and seasonal time evolution.…”

Section: Introductionmentioning

confidence: 99%

A Time Series Model Comparison for Monitoring and Forecasting Water Quality Variables

Monteiro

Costa

2018

Hydrology

View full text Add to dashboard Cite

Abstract:The monitoring and prediction of water quality parameters are important tasks in the management of water resources. In this work, the performances of time series statistical models were evaluated to predict and forecast the dissolved oxygen (DO) concentration in several monitoring sites located along the main river Vouga, in Portugal, during the period from January 2002 to May 2015. The models being compared are a regression model with correlated errors and a state-space model, which can be seen as a calibration model. Both models allow the incorporation of water quality variables, such as time correlation or seasonality. Results show that, for the DO variable, the calibration model outperforms the regression model for sample modeling, that is, for a short-term forecast, while the regression model with correlated errors has a better performance for the forecasting h-steps ahead framework. So, the calibration model is more useful for water monitoring using an online or real-time procedure, while the regression model with correlated errors can be applied in order to forecast over a longer period of time.

show abstract

“…an aggregated multiwavelet forecast without treating their models probabilistically nor studying the reliability of the forecasts (such as in Alizadeh, Jafari Nodoushan, et al (2017); Barzegar, Asghari Moghaddam, Adamowski, et al (2018), Barzegar et al (2017)).…”

Section: Water Resources Researchmentioning

confidence: 99%

A Stochastic Data‐Driven Ensemble Forecasting Framework for Water Resources: A Case Study Using Ensemble Members Derived From a Database of Deterministic Wavelet‐Based Models

Quilty

Adamowski

Boucher

2019

Water Resources Research

Self Cite

View full text Add to dashboard Cite

In water resources applications (e.g., streamflow, rainfall‐runoff, urban water demand [UWD], etc.), ensemble member selection and ensemble member weighting are two difficult yet important tasks in the development of ensemble forecasting systems. We propose and test a stochastic data‐driven ensemble forecasting framework that uses archived deterministic forecasts as input and results in probabilistic water resources forecasts. In addition to input data and (ensemble) model output uncertainty, the proposed approach integrates both ensemble member selection and weighting uncertainties, using input variable selection and data‐driven methods, respectively. Therefore, it does not require one to perform ensemble member selection and weighting separately. We applied the proposed forecasting framework to a previous real‐world case study in Montreal, Canada, to forecast daily UWD at multiple lead times. Using wavelet‐based forecasts as input data, we develop the Ensemble Wavelet‐Stochastic Data‐Driven Forecasting Framework, the first multiwavelet ensemble stochastic forecasting framework that produces probabilistic forecasts. For the considered case study, several variants of Ensemble Wavelet‐Stochastic Data‐Driven Forecasting Framework, produced using different input variable selection methods (partial correlation input selection and Edgeworth Approximations‐based conditional mutual information) and data‐driven models (multiple linear regression, extreme learning machines, and second‐order Volterra series models), are shown to outperform wavelet‐ and nonwavelet‐based benchmarks, especially during a heat wave (first time studied in the UWD forecasting literature).

show abstract

Multi-step water quality forecasting using a boosting ensemble multi-wavelet extreme learning machine model

Cited by 100 publications

References 51 publications

Application of SARIMA model to forecasting monthly flows in Waterval River, South Africa

Application of SARIMA model to forecasting monthly flows in Waterval River, South Africa

A Time Series Model Comparison for Monitoring and Forecasting Water Quality Variables

A Stochastic Data‐Driven Ensemble Forecasting Framework for Water Resources: A Case Study Using Ensemble Members Derived From a Database of Deterministic Wavelet‐Based Models

Contact Info

Product

Resources

About