Streamflow forecasting using extreme gradient boosting model coupled with Gaussian mixture model

Ni, Lingling; Wang, Dong; Wu, Jianfeng; Wang, Yuankun; Tao, Yong; Zhang, Jianyun; Liu, Jiufu

doi:10.1016/j.jhydrol.2020.124901

Cited by 145 publications

(49 citation statements)

References 50 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…Its urbanised landscape has also contributed to periodic historical flooding (Toronto and Region Conservation Authority, 2020a). Persistent severe flooding (recently in 2005 and 2013) has motivated calls for further mitigation strategies such as improved flow forecast models and early warning systems (Nirupama et al, 2014).…”

Section: Study Areamentioning

confidence: 99%

“…Gradient-boosting algorithms have previously been used to improve efficiency and accuracy for hydrological forecasting applications. Ni et al (2020) use the gradient-boosting variant XGBoost, which uses decision trees (DTs) as the individual learners in combination with a Gaussian mixture model (GMM) for streamflow forecasting. The GMM is used to cluster streamflow data, and an XGBoost ensemble is fitted to each cluster.…”

Section: Least-squares Boostingmentioning

confidence: 99%

See 1 more Smart Citation

Resampling and ensemble techniques for improving ANN-based high-flow forecast accuracy

Snieder

Abogadil

Khan

2021

Hydrol. Earth Syst. Sci.

View full text Add to dashboard Cite

Abstract. Data-driven flow-forecasting models, such as artificial neural networks (ANNs), are increasingly featured in research for their potential use in operational riverine flood warning systems. However, the distributions of observed flow data are imbalanced, resulting in poor prediction accuracy on high flows in terms of both amplitude and timing error. Resampling and ensemble techniques have been shown to improve model performance on imbalanced datasets. However, the efficacy of these methods (individually or combined) has not been explicitly evaluated for improving high-flow forecasts. In this research, we systematically evaluate and compare three resampling methods, random undersampling (RUS), random oversampling (ROS), and the synthetic minority oversampling technique for regression (SMOTER), and four ensemble techniques, randomised weights and biases, bagging, adaptive boosting (AdaBoost), and least-squares boosting (LSBoost), on their ability to improve high stage prediction accuracy using ANNs. These methods are implemented both independently and in combined hybrid techniques, where the resampling methods are embedded within the ensemble methods. This systematic approach for embedding resampling methods is a novel contribution. This research presents the first analysis of the effects of combining these methods on high stage prediction accuracy. Data from two Canadian watersheds (the Bow River in Alberta and the Don River in Ontario), representing distinct hydrological systems, are used as the basis for the comparison of the methods. The models are evaluated on overall performance and on typical and high stage subsets. The results of this research indicate that resampling produces marginal improvements to high stage prediction accuracy, whereas ensemble methods produce more substantial improvements, with or without resampling. Many of the techniques used produced an asymmetric trade-off between typical and high stage performance; reduction of high stage error resulted in disproportionately larger error on a typical stage. The methods proposed in this study highlight the diversity-in-learning concept and help support future studies on adapting ensemble algorithms for resampling. This research contains many of the first instances of such methods for flow forecasting and, moreover, their efficacy in addressing the imbalance problem and heteroscedasticity, which are commonly observed in high-flow and flood-forecasting models.

show abstract

Section: Study Areamentioning

confidence: 99%

Section: Least-squares Boostingmentioning

confidence: 99%

Resampling and ensemble techniques for improving ANN-based high-flow forecast accuracy

Snieder

Abogadil

Khan

2021

Hydrol. Earth Syst. Sci.

View full text Add to dashboard Cite

show abstract

“…These models can be classified into two types. The first type includes process-based models in which a hydrological model is established to calculate the response of rainfall based on empirical equations (conceptual) or physical processes (Ni et al 2020). These require the calibration of parameters to estimate, for example, evapotranspiration, infiltration, percolation, surface runoff, and other processes occurring in a basin (Kumanlioglu & Fistikoglu 2019).…”

mentioning

confidence: 99%

Untangling hybrid hydrological models with explainable artificial intelligence

2021

View full text Add to dashboard Cite

Hydrological models are valuable tools for developing streamflow predictions in unmonitored catchments to increase our understanding of hydrological processes. A recent effort has been made in the development of hybrid (conceptual/machine learning) models that can preserve some of the hydrological processes represented by conceptual models and can improve streamflow predictions. However, these studies have not explored how the data-driven component of hybrid models resolved runoff routing. In this study, explainable artificial intelligence (XAI) techniques are used to turn a ‘black-box’ model into a ‘glass box’ model. The hybrid models reduced the root-mean-square error of the simulated streamflow values by approximately 27, 50, and 24% for stations 17120000, 27380000, and 33680000, respectively, relative to the traditional method. XAI techniques helped unveil the importance of accounting for soil moisture in hydrological models. Differing from purely data-driven hydrological models, the inclusion of the production storage in the proposed hybrid model, which is responsible for estimating the water balance, reduced the short- and long-term dependencies of input variables for streamflow prediction. In addition, soil moisture controlled water percolation, which was the main predictor of streamflow. This finding is because soil moisture controls the underlying mechanisms of groundwater flow into river streams.

show abstract

“…The neuron has two gates, simplifying the structure of LSTM: an update gate and a reset gate, which modulate the flow of information when the neural state is updated at each step. The update gate decides how much information from the previous step flows to the neuron and the reset gate controls whether to ignore the previous state and upset the current or not [17,32]. This allows the neuron to drop irrelevant information, which is not useful for mapping the input and output in the future and further helps the neuron remember long-term information by reducing the memory burden [13].…”

Section: Methodology 21 Grumentioning

confidence: 99%

“…To further enhance forecast accuracy, hybrid models are becoming popular alternatives because they can outperform single models [17]. Combining methods can build hybrid models with optimization algorithms, which can be used to identify the optimal parameter combination and calibrate models to enhance their robustness.…”

Section: Introductionmentioning

confidence: 99%

Streamflow Forecasting via Two Types of Predictive Structure-Based Gated Recurrent Unit Models

Zhao

Wei

et al. 2021

Water

View full text Add to dashboard Cite

Data-intelligent methods designed for forecasting the streamflow of the Fenhe River are crucial for enhancing water resource management. Herein, the gated recurrent unit (GRU) is coupled with the optimization algorithm improved grey wolf optimizer (IGWO) to design a hybrid model (IGWO-GRU) to carry out streamflow forecasting. Two types of predictive structure-based models (sequential IGWO-GRU and monthly IGWO-GRU) are compared with other models, such as the single least-squares support vector machine (LSSVM) and single extreme learning machine (ELM) models. These models incorporate the historical streamflow series as inputs of the model to forecast the future streamflow with data from January 1956 to December 2016 at the Shangjingyou station and from January 1958 to December 2016 at the Fenhe reservoir station. The IGWO-GRU model exhibited a strong ability for mapping in streamflow series when the parameters were carefully tuned. The monthly predictive structure can effectively extract the instinctive hydrological information that is more easily learned by the predictive model than the traditional sequential predictive structure. The monthly IGWO-GRU model was found to be a better forecasting tool, with an average qualification rate of 91.66% in two stations. It also showed good performance in absolute error and peak flow forecasting.

show abstract

Streamflow forecasting using extreme gradient boosting model coupled with Gaussian mixture model

Cited by 145 publications

References 50 publications

Resampling and ensemble techniques for improving ANN-based high-flow forecast accuracy

Resampling and ensemble techniques for improving ANN-based high-flow forecast accuracy

Untangling hybrid hydrological models with explainable artificial intelligence

Streamflow Forecasting via Two Types of Predictive Structure-Based Gated Recurrent Unit Models

Contact Info

Product

Resources

About