A new approach to river flow forecasting: LSTM and GRU multivariate models

Melo, Gabriel Adriano; Sugimoto, Dylan Nakandakari; Tasinaffo, Paulo Marcelo; Santos, Afonso Henriques Moreira; Cunha, Adilson Marques da; Dias, Luiz Alberto Vieira

doi:10.1109/tla.2019.9011542

Cited by 19 publications

(7 citation statements)

References 11 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…These methods can improve the prediction accuracy by using different models; however, such methods are not designed for time series data, and do not consider the characteristics of streamflow. In addition, some time series methods which have been to for streamflow prediction includes Long-Short Term Memory (LSTM) [26,27] and Gated Recurrent Unit (GRU) [28], a combined model of feed-forward neural network (FNN) with particle swarm optimization (PSO) and gravitational search algorithm (GSA) [29]. Here, LSTM is a special kind of RNN developed to avoid the long-term dependency problem, which makes it hard to learn dependencies over long time windows [30].…”

Section: Introductionmentioning

confidence: 99%

Prediction of Streamflow Based on Dynamic Sliding Window LSTM

Dong

Fang

Wang

et al. 2020

Water

View full text Add to dashboard Cite

The streamflow of the upper reaches of the Yangtze River exhibits different timing and periodicity characteristics in different quarters and months of the year, which makes it difficult to predict. Existing sliding window-based methods usually use a fixed-size window, for which the window size selection is random, resulting in large errors. This paper proposes a dynamic sliding window method that reflects the different timing and periodicity characteristics of the streamflow in different months of the year. Multiple datasets of different months are generated using a dynamic window at first, then the long-short term memory (LSTM) is used to select the optimal window, and finally, the dataset of the optimal window size is used for verification. The proposed method was tested using the hydrological data of Zhutuo Hydrological Station (China). A comparison between the flow prediction data and the measured data shows that the prediction method based on a dynamic sliding window LSTM is more accurate by 8.63% and 3.85% than the prediction method based on fixed window LSTM and the dynamic sliding window back-propagation neural network, respectively. This method can be generally used for the time series data prediction with different periodic characteristics.

show abstract

Section: Introductionmentioning

confidence: 99%

Prediction of Streamflow Based on Dynamic Sliding Window LSTM

Dong

Fang

Wang

et al. 2020

Water

View full text Add to dashboard Cite

show abstract

“…extract before and after the relevance of data, is a variant of RNN that can overcome the weight exponential explosion or disappearance phenomenon and the problem of difficult network convergence by importing the gate control mechanism. It is widely applied in sequence data processing [32,42]. LSTM network is composed of several LSTM memory units in chronological order and the structure is shown in Fig.…”

Section: Long Short-term Memory Lstm Which Canmentioning

confidence: 99%

“…LSTM part: this part includes three LSTM layers, which sizes are (12,120), (12,60) and (12,32), respectively, and two dense layers. After learning by LSTM part, the data will be used as input for the next part.…”

Section: Lstm-attention Prediction Modulementioning

confidence: 99%

“…The informer method [27], which is proposed by Haoyi Zhou, et al, can capture the long-range dependency coupling between the input and output data so that the method is more suitable for the time sequence of very long-time spans and have some applications in other fields [28][29][30]. The different neural network algorithms are mutually combined with each other to improve forecasting accuracy, such as CNN-LSTM [31], LSTM-GRU [32] and GRU-CNN [33]. Some researchers studied the comparison with non-deep Learning, Deep Learning, and Ensemble Learning Methods [34].…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

The Sunspot Number Forecasting Using a Hybridization Model of EMD, LSTM and Attention Mechanism

Yang,

Fu,

Chen

2023

IEEJ Transactions Elec Engng

View full text Add to dashboard Cite

Sunspot number forecasting is a significant task for human beings in order to observe solar activity, and it is a classical chaotic time series. To improve the forecasting accuracy, we proposed a hybrid forecasting model based on empirical mode decomposition, long short‐term memory neural network and attention mechanism. First, the empirical mode decomposition is used to transfer the sunspot number to several sub‐components called IMFs and residual. Then, the IMFs and residual are input to the prediction module that is composed of long short‐term memory neural network and attention mechanism. Finally, the forecasting result is accumulated by the IMFs and residual through the predict module. The RMSE, MAE, MAPE and R2 indexes of our method are 1.567, 1.048, 1.8% and 0.9997. We also compare with GRU, informer and XGBoost‐DL, our method improves accuracy above 24 and 18 of RMSE and MAE, respectively. The experiment results demonstrate our method has higher precision. © 2023 Institute of Electrical Engineer of Japan and Wiley Periodicals LLC.

show abstract

“…Due to the ability of neural networks to mine the nonlinear relationships and deep features in training data, they generally have better forecasting performance [11,12], and are now widely used in forecasting for wind power generation [13]. Comparisons in the literature [14][15][16][17] between LSTM (Long Short-Term Memory neural network) and other forecasting models have shown LSTM models to be better than other forecasting models for both long-term and short-term forecasting; however, the LSTM model has the problems of model complexity and long training time. Thus, proper data processing can enhance the learning effect of the LSTM model.…”

Section: Introductionmentioning

confidence: 99%

Wind Power Forecasting Based on LSTM Improved by EMD-PCA-RF

2022

View full text Add to dashboard Cite

Improving the accuracy of wind power forecasting can guarantee the stable dispatch and safe operation of the grid system. Here, we propose an EMD-PCA-RF-LSTM wind power forecasting model to solve problems in traditional wind power forecasting such as incomplete consideration of influencing factors, inaccurate feature identification, and complex space–time relationships between variables. The proposed model incorporates Empirical Mode Decomposition (EMD), Principal Component Analysis (PCA), Random Forest (RF), and Long Short-Term Memory (LSTM) neural networks, And environmental factors are filtered by the Least Absolute Shrinkage and Selection Operator (LASSO) algorithm when pre-processing the data. First, the environmental factors are extended by the EMD algorithm to reduce the non-stationarity of the series. Second, the key influence series are extracted by the PCA algorithm in order to remove noisy information, which can seriously interfere with the data regression analysis. The data are then subjected to further feature extraction by calculating feature importance through the RF algorithm. Finally, the LSTM algorithm is used to perform dynamic time modeling of multivariate feature series for wind power forecasting. The above combined model is beneficial for analyzing the effects of different environmental factors on wind power and for obtaining more accurate prediction results. In a case study, the proposed combined forecasting model was verified using actual measured data from a power station. The results indicate that the proposed model provides the most accurate results when compared to benchmark models: MSE 7.26711 MW, RMSE 2.69576 MW, MAE 1.73981 MW, and adj-R2 0.9699203s.

show abstract

A new approach to river flow forecasting: LSTM and GRU multivariate models

Cited by 19 publications

References 11 publications

Prediction of Streamflow Based on Dynamic Sliding Window LSTM

Prediction of Streamflow Based on Dynamic Sliding Window LSTM

The Sunspot Number Forecasting Using a Hybridization Model of EMD, LSTM and Attention Mechanism

Wind Power Forecasting Based on LSTM Improved by EMD-PCA-RF

Contact Info

Product

Resources

About