Forecasting Day-Ahead Hourly Photovoltaic Power Generation Using Convolutional Self-Attention Based Long Short-Term Memory

The problem of Photovoltaic (PV) power generation forecasting is becoming crucial as the penetration level of Distributed Energy Resources (DERs) increases in microgrids and Virtual Power Plants (VPPs). In order to improve the stability of power systems, a fair amount of research has been proposed for increasing prediction performance in practical environments through statistical, machine learning, deep learning, and hybrid approaches. Despite these efforts, the problem of forecasting PV power generation remains to be challenging in power system operations since existing methods show limited accuracy and thus are not sufficiently practical enough to be widely deployed. Many existing methods using long historical data suffer from the long-term dependency problem and are not able to produce high prediction accuracy due to their failure to fully utilize all features of long sequence inputs. To address this problem, we propose a deep learning-based PV power generation forecasting model called Convolutional Self-Attention based Long Short-Term Memory (LSTM). By using the convolutional self-attention mechanism, we can significantly improve prediction accuracy by capturing the local context of the data and generating keys and queries that fit the local context. To validate the applicability of the proposed model, we conduct extensive experiments on both PV power generation forecasting using a real world dataset and power consumption forecasting. The experimental results of power generation forecasting using the real world datasets show that the MAPEs of the proposed model are much lower, in fact by 7.7%, 6%, 3.9% compared to the Deep Neural Network (DNN), LSTM and LSTM with the canonical self-attention, respectively. As for power consumption forecasting, the proposed model exhibits 32%, 17% and 44% lower Mean Absolute Percentage Error (MAPE) than the DNN, LSTM and LSTM with the canonical self-attention, respectively.

Download Full-text

DLI: A Deep Learning-Based Granger Causality Inference

Complexity ◽

10.1155/2020/5960171 ◽

2020 ◽

Vol 2020 ◽

pp. 1-6 ◽

Cited By ~ 3

Author(s):

Wei Peng

Keyword(s):

Deep Learning ◽

Granger Causality ◽

Prediction Accuracy ◽

Short Term Memory ◽

Target Variable ◽

Short Term ◽

Term Memory ◽

P Index ◽

Causality Inference ◽

Long Short Term Memory

Integrating autoencoder (AE), long short-term memory (LSTM), and convolutional neural network (CNN), we propose an interpretable deep learning architecture for Granger causality inference, named deep learning-based Granger causality inference (DLI). Two contributions of the proposed DLI are to reveal the Granger causality between the bitcoin price and S&P index and to forecast the bitcoin price and S&P index with a higher accuracy. Experimental results demonstrate that there is a bidirectional but asymmetric Granger causality between the bitcoin price and S&P index. And the DLI performs a superior prediction accuracy by integrating variables that have causalities with the target variable into the prediction process.

Download Full-text

CONVOLUTIONAL LONG-SHORT TERM MEMORY NETWORKS MODEL FOR LONG DURATION EEG SIGNAL CLASSIFICATION

Journal of Mechanics in Medicine and Biology ◽

10.1142/s0219519419400050 ◽

2019 ◽

Vol 19 (01) ◽

pp. 1940005 ◽

Cited By ~ 3

Author(s):

ULAS BARAN BALOGLU ◽

ÖZAL YILDIRIM

Keyword(s):

Deep Learning ◽

Short Term Memory ◽

Learning Algorithms ◽

Eeg Signals ◽

Short Term ◽

Term Memory ◽

Long Duration ◽

Proposed Model ◽

Clinical Scenarios ◽

Long Short Term Memory

Background and objective: Deep learning structures have recently achieved remarkable success in the field of machine learning. Convolutional neural networks (CNN) in image processing and long-short term memory (LSTM) in the time-series analysis are commonly used deep learning algorithms. Healthcare applications of deep learning algorithms provide important contributions for computer-aided diagnosis research. In this study, convolutional long-short term memory (CLSTM) network was used for automatic classification of EEG signals and automatic seizure detection. Methods: A new nine-layer deep network model consisting of convolutional and LSTM layers was designed. The signals processed in the convolutional layers were given as an input to the LSTM network whose outputs were processed in densely connected neural network layers. The EEG data is appropriate for a model having 1-D convolution layers. A bidirectional model was employed in the LSTM layer. Results: Bonn University EEG database with five different datasets was used for experimental studies. In this database, each dataset contains 23.6[Formula: see text]s duration 100 single channel EEG segments which consist of 4097 dimensional samples (173.61[Formula: see text]Hz). Eight two-class and three three-class clinical scenarios were examined. When the experimental results were evaluated, it was seen that the proposed model had high accuracy on both binary and ternary classification tasks. Conclusions: The proposed end-to-end learning structure showed a good performance without using any hand-crafted feature extraction or shallow classifiers to detect the seizures. The model does not require filtering, and also automatically learns to filter the input as well. As a result, the proposed model can process long duration EEG signals without applying segmentation, and can detect epileptic seizures automatically by using the correlation of ictal and interictal signals of raw data.

Download Full-text

Attention-Based Long Short-Term Memory Recurrent Neural Network for Capacity Degradation of Lithium-Ion Batteries

Batteries ◽

10.3390/batteries7040066 ◽

2021 ◽

Vol 7 (4) ◽

pp. 66

Author(s):

Tadele Mamo ◽

Fu-Kwun Wang

Keyword(s):

Neural Network ◽

Recurrent Neural Network ◽

Prediction Accuracy ◽

Short Term Memory ◽

Lithium Ion ◽

Short Term ◽

Term Memory ◽

Proposed Model ◽

Capacity Degradation ◽

Long Short Term Memory

Monitoring cycle life can provide a prediction of the remaining battery life. To improve the prediction accuracy of lithium-ion battery capacity degradation, we propose a hybrid long short-term memory recurrent neural network model with an attention mechanism. The hyper-parameters of the proposed model are also optimized by a differential evolution algorithm. Using public battery datasets, the proposed model is compared to some published models, and it gives better prediction performance in terms of mean absolute percentage error and root mean square error. In addition, the proposed model can achieve higher prediction accuracy of battery end of life.

Download Full-text

A Hybrid Model for Short-Term Traffic Flow Prediction Based on Variational Mode Decomposition, Wavelet Threshold Denoising, and Long Short-Term Memory Neural Network

Complexity ◽

10.1155/2021/7756299 ◽

2021 ◽

Vol 2021 ◽

pp. 1-24

Author(s):

Yang Yu ◽

Qiang Shang ◽

Tian Xie

Keyword(s):

Traffic Flow ◽

Prediction Accuracy ◽

Short Term Memory ◽

Short Term ◽

Term Memory ◽

Traffic Flow Prediction ◽

Flow Prediction ◽

Mode Decomposition ◽

Proposed Model ◽

Long Short Term Memory

Traffic flow prediction plays an important role in intelligent transportation system (ITS). However, due to the randomness and complex periodicity of traffic flow data, traditional prediction models often fail to achieve good results. On the other hand, external disturbances or abnormal detectors will cause the collected traffic flow data to contain noise components, resulting in a decrease in prediction accuracy. In order to improve the accuracy of traffic flow prediction, this study proposes a mixed traffic flow prediction model VMD-WD-LSTM using variational mode decomposition (VMD), wavelet threshold denoising (WD), and long short-term memory (LSTM) network. Firstly, we decompose the original traffic flow sequence into K components through VMD and determine the number of components K according to the sample entropy of different K values. Then, each component is denoised by wavelet threshold to obtain the denoised subsequence. Finally, LSTM is used to predict each subsequence, and the predicted values of each subsequence are combined into the final prediction results. In addition, the performance of the proposed model and the latest traffic flow prediction model is compared on the several well-known public datasets. The empirical analysis shows that the proposed model not only has good prediction accuracy but also has superior robustness.

Download Full-text

Classification and Prediction of Typhoon Levels by Satellite Cloud Pictures through GC–LSTM Deep Learning Model

Sensors ◽

10.3390/s20185132 ◽

2020 ◽

Vol 20 (18) ◽

pp. 5132

Author(s):

Jianyin Zhou ◽

Jie Xiang ◽

Sixun Huang

Keyword(s):

Deep Learning ◽

Prediction Accuracy ◽

Prediction Models ◽

Short Term Memory ◽

Network Graph ◽

Short Term ◽

Convolutional Network ◽

Term Memory ◽

Model Stability ◽

Long Short Term Memory

Typhoons are some of the most serious natural disasters, and the key to disaster prevention and mitigation is typhoon level classification. How to better use data of satellite cloud pictures to achieve accurate classification of typhoon levels has become one of classification the hot issues in current studies. A new framework of deep learning neural network, Graph Convolutional–Long Short-Term Memory Network (GC–LSTM), is proposed, which is based on the data of satellite cloud pictures of the Himawari-8 satellite in 2010–2019. The Graph Convolutional Network (GCN) is used to process the irregular spatial structure of satellite cloud pictures effectively, and the Long Short-Term Memory (LSTM) network is utilized to learn the characteristics of satellite cloud pictures over time. Moreover, to verify the effectiveness and accuracy of the model, the prediction effect and model stability are compared with other models. The results show that: the algorithm performance of this model is better than other prediction models; the prediction accuracy rate of typhoon level classification reaches 92.35%, and the prediction accuracy of typhoons and super typhoons reaches 95.12%. The model can accurately identify typhoon eye and spiral cloud belt, and the prediction results are always kept in the minimum range compared with the actual results, which proves that the GC–LSTM model has stronger stability. The model can accurately identify the levels of different typhoons according to the satellite cloud pictures. In summary, the results can provide a theoretical basis for the related research of typhoon level classification.

Download Full-text

Non-Intrusive Load Identification Method Based on Improved Long Short Term Memory Network

Energies ◽

10.3390/en14030684 ◽

2021 ◽

Vol 14 (3) ◽

pp. 684

Author(s):

Jiateng Song ◽

Hongbin Wang ◽

Mingxing Du ◽

Lei Peng ◽

Shuai Zhang ◽

...

Keyword(s):

Deep Learning ◽

Short Term Memory ◽

Rapid Development ◽

Series Data ◽

Load Identification ◽

Short Term ◽

Term Memory ◽

Identification Method ◽

Proposed Model ◽

Long Short Term Memory

Non-intrusive load monitoring (NILM) is an important research direction and development goal on the distribution side of smart grid, which can significantly improve the timeliness of demand side response and users’ awareness of load. Due to rapid development, deep learning becomes an effective way to optimize NILM. In this paper, we propose a novel load identification method based on long short term memory (LSTM) on deep learning. Sequence-to-point (seq2point) learning is introduced into LSTM. The innovative combination of the LSTM and the seq2point brings their respective advantages together, so that the proposed model can accurately identify the load in process of time series data. In this paper, we proved the feature of reducing identification error in the experimental data, from three datasets, UK-DALE dataset, REDD dataset, and REFIT dataset. In terms of mean absolute error (MAE), the three datasets have increased by 15%, 14%, and 18% respectively; in terms of normalized signal aggregate error (SAE), the three datasets have increased by 21%, 24%, and 30% respectively. Compared with the existing models, the proposed model has better accuracy and generalization in identifying three open source datasets.

Download Full-text

A COMBINED DEEP LEARNING MODEL FOR PERSIAN SENTIMENT ANALYSIS

IIUM Engineering Journal ◽

10.31436/iiumej.v20i1.1036 ◽

2019 ◽

Vol 20 (1) ◽

pp. 129-139 ◽

Cited By ~ 2

Author(s):

Zahra Bokaee Nezhad ◽

Mohammad Ali Deihimi

Keyword(s):

Deep Learning ◽

Natural Language Processing ◽

Sentiment Analysis ◽

Language Processing ◽

Short Term Memory ◽

Short Term ◽

Term Memory ◽

Proposed Model ◽

Long Short Term Memory ◽

Deep Learning Model

With increasing members in social media sites today, people tend to share their views about everything online. It is a convenient way to convey their messages to end users on a specific subject. Sentiment Analysis is a subfield of Natural Language Processing (NLP) that refers to the identification of users’ opinions toward specific topics. It is used in several fields such as marketing, customer services, etc. However, limited works have been done on Persian Sentiment Analysis. On the other hand, deep learning has recently become popular because of its successful role in several Natural Language Processing tasks. The objective of this paper is to propose a novel hybrid deep learning architecture for Persian Sentiment Analysis. According to the proposed model, local features are extracted by Convolutional Neural Networks (CNN) and long-term dependencies are learned by Long Short Term Memory (LSTM). Therefore, the model can harness both CNN's and LSTM's abilities. Furthermore, Word2vec is used for word representation as an unsupervised learning step. To the best of our knowledge, this is the first attempt where a hybrid deep learning model is used for Persian Sentiment Analysis. We evaluate the model on a Persian dataset that is introduced in this study. The experimental results show the effectiveness of the proposed model with an accuracy of 85%. ABSTRAK: Hari ini dengan ahli yang semakin meningkat di laman media sosial, orang cenderung untuk berkongsi pandangan mereka tentang segala-galanya dalam talian. Ini adalah cara mudah untuk menyampaikan mesej mereka kepada pengguna akhir mengenai subjek tertentu. Analisis Sentimen adalah subfield Pemprosesan Bahasa Semula Jadi yang merujuk kepada pengenalan pendapat pengguna ke arah topik tertentu. Ia digunakan dalam beberapa bidang seperti pemasaran, perkhidmatan pelanggan, dan sebagainya. Walau bagaimanapun, kerja-kerja terhad telah dilakukan ke atas Analisis Sentimen Parsi. Sebaliknya, pembelajaran mendalam baru menjadi popular kerana peranannya yang berjaya dalam beberapa tugas Pemprosesan Bahasa Asli (NLP). Objektif makalah ini adalah mencadangkan senibina pembelajaran hibrid yang baru dalam Analisis Sentimen Parsi. Menurut model yang dicadangkan, ciri-ciri tempatan ditangkap oleh Rangkaian Neural Convolutional (CNN) dan ketergantungan jangka panjang dipelajari oleh Long Short Term Memory (LSTM). Oleh itu, model boleh memanfaatkan kebolehan CNN dan LSTM. Selain itu, Word2vec digunakan untuk perwakilan perkataan sebagai langkah pembelajaran tanpa pengawasan. Untuk pengetahuan yang terbaik, ini adalah percubaan pertama di mana model pembelajaran mendalam hibrid digunakan untuk Analisis Sentimen Persia. Kami menilai model pada dataset Persia yang memperkenalkan dalam kajian ini. Keputusan eksperimen menunjukkan keberkesanan model yang dicadangkan dengan ketepatan 85%.

Download Full-text

Entity-Extraction Using Hybrid Deep-Learning Approach for Hindi text

International Journal of Cognitive Informatics and Natural Intelligence ◽

10.4018/ijcini.20210701.oa1 ◽

2021 ◽

Vol 15 (3) ◽

pp. 1-11

Author(s):

Richa Sharma ◽

Sudha Morwal ◽

Basant Agarwal

Keyword(s):

Neural Network ◽

Deep Learning ◽

Short Term Memory ◽

Conditional Random Field ◽

Research Work ◽

Entity Recognition ◽

Short Term ◽

Term Memory ◽

Proposed Model ◽

Long Short Term Memory

This article presents a neural network-based approach to develop named entity recognition for Hindi text. In this paper, the authors propose a deep learning architecture based on convolutional neural network (CNN) and bi-directional long short-term memory (Bi-LSTM) neural network. Skip-gram approach of word2vec model is used in the proposed model to generate word vectors. In this research work, several deep learning models have been developed and evaluated as baseline systems such as recurrent neural network (RNN), long short-term memory (LSTM), Bi-LSTM. Furthermore, these baseline systems are promoted to a proposed model with the integration of CNN and conditional random field (CRF) layers. After a comparative analysis of results, it is verified that the performance of the proposed model (i.e., Bi-LSTM-CNN-CRF) is impressive. The proposed system achieves 61% precision, 56% recall, and 58% F-measure.

Download Full-text

Container Volume Prediction Using Time-Series Decomposition with a Long Short-Term Memory Models

Applied Sciences ◽

10.3390/app11198995 ◽

2021 ◽

Vol 11 (19) ◽

pp. 8995

Author(s):

Eunju Lee ◽

Dohee Kim ◽

Hyerim Bae

Keyword(s):

Time Series ◽

Deep Learning ◽

Prediction Models ◽

Short Term Memory ◽

Short Term ◽

Term Memory ◽

Proposed Model ◽

External Variables ◽

Long Short Term Memory ◽

Better Than

The purpose of this study is to improve the prediction of container volumes in Busan ports by applying external variables and time-series data decomposition methods to deep learning prediction models. Previous studies on container volume forecasting were based on traditional statistical methodologies, such as ARIMA, SARIMA, and regression. However, these methods do not explain the complexity and variability of data caused by changes in the external environment, such as the global financial crisis and economic fluctuations. Deep learning can explore the inherent patterns of data and analyze the characteristics (time series, external environmental variables, and outliers); hence, the accuracy of deep learning-based volume prediction models is better than that of traditional models. However, this does not include the study of overall trends (upward, steady, or downward). In this study, a novel deep learning prediction model is proposed that combines prediction and trend identification of container volume. The proposed model explores external variables that are related to container volume, combining port volume time-series decomposition with external variables and deep learning-based multivariate long short-term memory (LSTM) prediction. The results indicate that the proposed model performs better than the traditional LSTM model and follows the trend simultaneously.

Download Full-text

Deep Learning with Long Short Term Memory Based Sequence-to-Sequence Model for Rainfall-Runoff Simulation

Water ◽

10.3390/w13040437 ◽

2021 ◽

Vol 13 (4) ◽

pp. 437

Author(s):

Heechan Han ◽

Changhyun Choi ◽

Jaewon Jung ◽

Hung Soo Kim

Keyword(s):

Deep Learning ◽

Short Term Memory ◽

Support Vector ◽

Sequence Structure ◽

Short Term ◽

Term Memory ◽

Runoff Forecasting ◽

Proposed Model ◽

Runoff Prediction ◽

Long Short Term Memory

Accurate runoff prediction is one of the important tasks in various fields such as agriculture, hydrology, and environmental studies. Recently, with massive improvements of computational system and hardware, the deep learning-based approach has recently been applied for more accurate runoff prediction. In this study, the long short-term memory model with sequence-to-sequence structure was applied for hourly runoff predictions from 2015 to 2019 in the Russian River basin, California, USA. The proposed model was used to predict hourly runoff with lead time of 1–6 h using runoff data observed at upstream stations. The model was evaluated in terms of event-based performance using the statistical metrics including root mean square error, Nash-Sutcliffe Efficiency, peak runoff error, and peak time error. The results show that proposed model outperforms support vector machine and conventional long short-term memory models. In addition, the model has the best predictive ability for runoff events, which means that it can be effective for developing short-term flood forecasting and warning systems. The results of this study demonstrate that the deep learning-based approach for hourly runoff forecasting has high predictive power and sequence-to-sequence structure is effective method to improve the prediction results.

Download Full-text