A novel dynamic integration approach for multiple load forecasts based on Q‐learning algorithm

Ma, Minhua; Jin, Bingjie; Luo, Shuxin; Guo, Song; Huang, Hongwei

doi:10.1002/2050-7038.12146

Cited by 10 publications

(4 citation statements)

References 35 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…18 To counter these issues, GRU and LSTM based DL models are proposed which have some extra control gates, 18,19 these additional gates effectively addresses the gradient vanishing and the exploding problem of RNN. The DL models have used widely in different fields like: electrical load and power forecasting, 20,21 solar radiation and power forecasting, 14,[22][23][24] speech, 25 computer vision and text. 26,27 Hybrid DL algorithms are also analyzed like, in some works wavelet transform.…”

Section: Introductionmentioning

confidence: 99%

“…The CNN and LSTM based hybrid DL model is proposed for solar power forecasting with roughly estimated weather data. 21 Again in this work, standard errors are set as a benchmark for the prediction for short term solar power prediction.…”

Section: Introductionmentioning

confidence: 99%

“…However, in this work, hybrid DL models are not compared, and only standard errors are compared as a benchmark for comparison of proposed LSTM based prediction. The CNN and LSTM based hybrid DL model is proposed for solar power forecasting with roughly estimated weather data 21 . Again in this work, standard errors are set as a benchmark for the prediction for short term solar power prediction.…”

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations

A CNN‐BiLSTM based deep learning model for mid‐term solar radiation prediction

Rai

Shrivastava

Jana

2020

Int Trans Electr Energ Syst

View full text Add to dashboard Cite

The penetrations of solar power plants are increasing their presence worldwide.The solar power plants have uncertain power output as its output depends on solar radiation, which is environmental dependent, so solar radiation prediction is a crucial step in integrating these plants into the power grid. In this work, a convolution neural network (CNN) and bi-direction long short term memory (BiLSTM) based hybrid deep learning (DL) model is proposed for effective midterm solar radiation prediction. The CNN architecture in this model captures the feature in solar radiation input data, and BiLSTM exploits the dependencies of this time series data. The proposed model is tested for three different geographical locations on the same latitude as it receives approximately the same solar radiation. The proposed hybrid DL model is compared with different recently proposed DL models. Moreover, distribution errors such as, skew and kurtosis errors are included for evaluating the distribution of predicted solar radiation. The outcome shows that the proposed hybrid DL model is robust and further enhancing the recently proposed DL models for midterm solar radiation prediction.

show abstract

Section: Introductionmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

A CNN‐BiLSTM based deep learning model for mid‐term solar radiation prediction

Rai

Shrivastava

Jana

2020

Int Trans Electr Energ Syst

View full text Add to dashboard Cite

show abstract

“…This stochastic search algorithm is used to find the optimum number of neurons of the hidden layer. Ma et al 31 proposed a dynamic integration approach of multiple intelligent algorithms including neural network, deep learning, and so on. The method uses Q-learning algorithm to achieve adaptive selection of prediction methods and calculation of dynamic weights for different environments.…”

Section: Auto-regressive Movingmentioning

confidence: 99%

Short‐term load forecasting method based on deep neural network with sample weights

Cai

Yan

et al. 2020

Int Trans Electr Energ Syst

View full text Add to dashboard Cite

Summary The reform of power market has presented new challenges to short‐term load forecasting (STLF), and the accuracy of forecast results is of great significance to the orderly and efficient operation of the market. To improve the accuracy of load forecasting, an STLF method based on deep neural network (DNN) with sample weights is proposed. By filtering samples and assigning corresponding weights to different training samples, this method effectively improves the forecasting accuracy of the DNN model. Two important steps of this method are as follows: (a) calculating the distance between the samples and the forecast day to measure the similarity between them, and the training samples are selected based on the distance; (b) after selecting training samples, corresponding weights can be assigned to training samples according to the distance between the training samples and the forecast day. This allows DNN to focus on the key samples. Finally, we carried out simulation analysis by using the actual load data of Guangdong Power Grid in January 2017. The results show that the proposed method can effectively improve the accuracy and reliability of load forecasting.

show abstract