Long short-term memory-singular spectrum analysis-based model for electric load forecasting

Neeraj, Neeraj; Mathew, Jimson; Agarwal, Mayank; Behera, Ranjan Kumar

doi:10.1007/s00202-020-01135-y

Cited by 25 publications

(13 citation statements)

References 40 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…There are many precedents for the application of the signal processing method in time series forecasting. Some scientific fields include Neeraj et al [56], using the Single Spectrum Analysis Long-Short Term Memory (SSA-LSTM) model to forecast the power load, which achieves higher forecasting accuracy than most traditional methods in machine learning (Stylianos et al [57]). Using the Single Spectrum Analysis (SSA) with Artificial Neural networks (ANN) combined with the hybrid method to predict the time series of road traffic volume can improve the prediction accuracy of the traditional neural network.…”

Section: Related Workmentioning

confidence: 99%

Ensemble Empirical Mode Decomposition with Adaptive Noise with Convolution Based Gated Recurrent Neural Network: A New Deep Learning Model for South Asian High Intensity Forecasting

et al. 2021

View full text Add to dashboard Cite

The intensity variation of the South Asian high (SAH) plays an important role in the formation and extinction of many kinds of mesoscale systems, including tropical cyclones, southwest vortices in the Asian summer monsoon (ASM) region, and the precipitation in the whole Asia Europe region, and the SAH has a vortex symmetrical structure; its dynamic field also has the symmetry form. Not enough previous studies focus on the variation of SAH daily intensity. The purpose of this study is to establish a day-to-day prediction model of the SAH intensity, which can accurately predict not only the interannual variation but also the day-to-day variation of the SAH. Focusing on the summer period when the SAH is the strongest, this paper selects the geopotential height data between 1948 and 2020 from NCEP to construct the SAH intensity datasets. Compared with the classical deep learning methods of various kinds of efficient time series prediction model, we ultimately combine the Ensemble Empirical Mode Decomposition with Adaptive Noise (CEEMDAN) method, which has the ability to deal with the nonlinear and unstable single system, with the Permutation Entropy (PE) method, which can extract the SAH intensity feature of IMF decomposed by CEEMDAN, and the Convolution-based Gated Recurrent Neural Network (ConvGRU) model is used to train, test, and predict the intensity of the SAH. The prediction results show that the combination of CEEMDAN and ConvGRU can have a higher accuracy and more stable prediction ability than the traditional deep learning model. After removing the redundant features in the time series, the prediction accuracy of the SAH intensity is higher than that of the classical model, which proves that the method has good applicability for the prediction of nonlinear systems in the atmosphere.

show abstract

Section: Related Workmentioning

confidence: 99%

Ensemble Empirical Mode Decomposition with Adaptive Noise with Convolution Based Gated Recurrent Neural Network: A New Deep Learning Model for South Asian High Intensity Forecasting

et al. 2021

View full text Add to dashboard Cite

show abstract

“…Kalman filtering and auto regressive moving average model are used in [6] to significantly reduce the random noise in ultra‐short‐term power load forecast. Empirical mode decomposition (EMD) [7], singular spectrum analysis [8] are applied on LSTM model respectively, to train the modes extracted from load curves and observe irregular noise characteristics. However, problems such as mode mixing and so on are easy to occur during EMD decomposition, which led to the decrease of accuracy.…”

Section: Introductionmentioning

confidence: 99%

A Reliable Short‐Term Power Load Forecasting Method Based on VMD‐IWOA‐LSTM Algorithm

Zhuang

Zheng

Chen

et al. 2022

IEEJ Transactions Elec Engng

View full text Add to dashboard Cite

To reduce the short-term load forecasting (STLF) error of off-line forecasting model, a VMD-IWOA-LSTM (VIL) method for STLF is proposed. Firstly, variational mode decomposition (VMD) is used to decompose the historical power load signals. Then, the decomposed signals are reconstructed according to the similarity of Pearson correlation coefficient (PCC), and meteorological data are chosen for each reconstructed component based on the set PCC threshold. The long short-term memory (LSTM) models are used to predict the corresponding components, and improved whale optimization algorithm (IWOA) is used to optimize the parameters in LSTM. Finally, the forecast results of each component are added together to get the final forecast result. The experimental results of power load data in a certain area show that the proposed method has the advantages of strong anti-interference performance and high prediction accuracy compared with other methods, and has strong practicability.

show abstract

“…In previous research, the author developed the CNN-Bi-GRU to forecaste the load power of building, which achieved more improvement compared with other deep learning methodologies [34]. Neeraj et al developed the singular spectrum analysis LSTM for electric load forecasting [35]. e proposed method achieved superior forecasting performance in half-an-hour and one day-ahead compared with the existing method.…”

Section: Introductionmentioning

confidence: 99%

Three-Phase Load Prediction-Based Hybrid Convolution Neural Network Combined Bidirectional Long Short-Term Memory in Solar Power Plant

Lee

Thanh

Yeh

et al. 2022

International Transactions on Electrical Energy Systems

View full text Add to dashboard Cite

e economic renewable energy generations have been rapidly developed because of the sharp reduction in the costs of solar panels. It is imperative to forecast the three-phase load power for more effective energy planning and optimization in a smart solar microgrid installed on a building in the Linyuan District, Taiwan. To alleviate this problem, this article proposes a convolution neural network bidirectional long short-term memory (CNN-Bi-LSTM) to accurately predict the short-term three-phase load power in building the energy management system in the smart solar microgrid with the collected data from advanced metering infrastructure (AMI), which have not been investigated before. e three-phase load-predicting methodology is developed using weather parameters and different collected data from AMI. e project evaluates the performance of the CNN-Bi-LSTM model by utilizing hyper-parameter optimization to attain the optimum parameters. e prediction models are trained based on hourly historical input features, selected based on the Pearson correlation coefficient. e performances' optimal structure CNN-Bi-LSTM are validated and compared with the bidirectional LSTM (Bi-LSTM), LSTM, the Gated Recurrent Unit (GRU), and the recurrent neural network (RNN) models. e obtained optimized structure of CNN-Bi-LSTM demonstrates the effectiveness of the proposed models in the short-term prediction of three-phase load power in a smart solar microgrid for building with a maximum enhancement of 68.36% and 8.81% average MSE, and 30.26% and 36.36% average MAE during the testing and validating operations.

show abstract

Long short-term memory-singular spectrum analysis-based model for electric load forecasting

Cited by 25 publications

References 40 publications

Ensemble Empirical Mode Decomposition with Adaptive Noise with Convolution Based Gated Recurrent Neural Network: A New Deep Learning Model for South Asian High Intensity Forecasting

Ensemble Empirical Mode Decomposition with Adaptive Noise with Convolution Based Gated Recurrent Neural Network: A New Deep Learning Model for South Asian High Intensity Forecasting

A Reliable Short‐Term Power Load Forecasting Method Based on VMD‐IWOA‐LSTM Algorithm

Three-Phase Load Prediction-Based Hybrid Convolution Neural Network Combined Bidirectional Long Short-Term Memory in Solar Power Plant

Contact Info

Product

Resources

About