Short-term Neural Network Research Articles

Wind Power Short-Term Forecasting Based on LSTM Neural Network With Dragonfly Algorithm

The volatility and randomness of wind energy limit its large-scale usage in power systems. Accurate short-term wind power prediction can provide effective criteria for wind energy parallel in the grid and provide favorable conditions for the commercial utilization of wind energy. Therefore, the paper proposes a short-term wind power prediction model based on the dragonfly algorithm optimize long-term and short-term neural networks. Firstly, the model preprocesses the collected data and divides the data into a training set and a testing set. Then, the DA used the training set to optimize the relevant hyperparameters in the long and short-term memory neural network. Finally, the DA-LSTM prediction model constructed with excellent hyperparameters will use the test set to obtain the prediction results. The simulation results of the examples show that, compared with the GWO-BP, ELM, and LSTM models, the DA-LSTM model can effectively use time series data for short-term forecasting of wind power and has higher prediction accuracy.

Approach to COVID-19 time series data using deep learning and spectral analysis methods

<abstract> <p>This article focuses on the application of deep learning and spectral analysis to epidemiology time series data, which has recently piqued the interest of some researchers. The COVID-19 virus is still mutating, particularly the delta and omicron variants, which are known for their high level of contagiousness, but policymakers and governments are resolute in combating the pandemic's spread through a recent massive vaccination campaign of their population. We used extreme machine learning (ELM), multilayer perceptron (MLP), long short-term neural network (LSTM), gated recurrent unit (GRU), convolution neural network (CNN) and deep neural network (DNN) methods on time series data from the start of the pandemic in France, Russia, Turkey, India, United states of America (USA), Brazil and United Kingdom (UK) until September 3, 2021 to predict the daily new cases and daily deaths at different waves of the pandemic in countries considered while using root mean square error (RMSE) and relative root mean square error (rRMSE) to measure the performance of these methods. We used the spectral analysis method to convert time (days) to frequency in order to analyze the peaks of frequency and periodicity of the time series data. We also forecasted the future pandemic evolution by using ELM, MLP, and spectral analysis. Moreover, MLP achieved best performance for both daily new cases and deaths based on the evaluation metrics used. Furthermore, we discovered that errors for daily deaths are much lower than those for daily new cases. While the performance of models varies, prediction and forecasting during the period of vaccination and recent cases confirm the pandemic's prevalence level in the countries under consideration. Finally, some of the peaks observed in the time series data correspond with the proven pattern of weekly peaks that is unique to the COVID-19 time series data.</p> </abstract>

Neural Networks as Classification Mechanisms of Complex Human Activities

Within this paper, we present two neural nets for view-independent complex human activity recognition (HAR) from video frames. For our study here, we reduce the number of frames produced by a video sequence given that we can identify activities from a sparsely sampled sequence of body poses, and, at the same time, we are able to reduce the processing complexity and response while hardly affecting the accuracy, precision, and recall. To do so, we use a formal framework to ensure the quality of data collection and data preprocessing. We utilize neural networks for the classification of single and complex body activities. More specifically, we consider the sequence of body poses as a time-series problem given that they can provide state-of-the-art results on challenging recognition tasks with little data engineering. Deep Learning in the form of Convolutional Neural Network (CNN), Long Short-Term Neural Network (LSTM), and a one-dimensional Convolutional Neural Network Long Short-Term Memory model (CNN-LSTM) are used as benchmarks to classify the activity.

Stock Picking with Machine Learning

We combine insights from machine learning and finance research to build machine learn-ing algorithms for stock selection. Our study builds on weekly data for the historical constitu-ents of the S&P 500 over the period from January 1999 to March 2021 and includes typical equity factors as well as additional fundamental data, technical indicators, and historical re-turns. Deep neural networks (DNN), long short-term neural networks (LSTM), random forest, gradient boosting, and regularized logistic Regression models are trained on stock characteris-tics to predict whether a specific stock outperforms the market over the subsequent week. We analyze a trading strategy that picks stocks with the highest probability predictions to outper-form the market. Our empirical results show a substantial and significant outperformance of machine learning based stock selection models compared to a simple equally weighted bench-mark. Moreover, we find non-linear machine learning models such as neural networks and tree-based models to outperform more simple regularized logistic regression approaches. The re-sults are robust when applied to the STOXX Europe 600 as alternative asset universe. However, all analyzed machine learning strategies demonstrate a substantial portfolio turnover and trans-action costs have to be marginal to capitalize on the strategies.

LSTM-NN Yaw Control of Wind Turbines Based on Upstream Wind Information

Based on wind lidar, a novel yaw control scheme was designed that utilizes forecast wind information. The new scheme can reduce the power loss caused by the lag of accurate measurement data in the traditional yaw control strategy. A theoretical analysis of the power loss caused by the traditional wind measurement inherent error and the wind direction based traditional yaw control strategy was conducted. The yaw angle error and yaw stop/start frequency in an actual wind field were statistically analyzed, and a novel Long Short Term-Neural Network (LSTM-NN) yaw control strategy based on wind lidar information was proposed. An accurate forecast of the wind direction could reduce the power loss caused by the inherent yaw misalignment, while an accurate forecast of wind speed could increase the stop/start frequency in the medium speed section within the partial load range and reduce the frequency in the low speed section within the partial load range. Thus, the power captured could be increased by 3.5% under certain wind conditions without increasing the yaw duty. Based on a simple wind evolution model and a novel yaw control strategy, the validity of the yaw control strategy was verified in a FAST/Simulink simulation model.

Rainfall Monitoring Based on Machine Learning by Earth-Space Link in the Ku Band

Recently, the oblique earth-space links (OELs) between satellite and earth station have been used for rainfall monitoring as a supplement to existing observation methods. Most recent studies achieved the rainfall measurement by OELs based on the empirical method such as power-law (PL) model. In practice, two crucial issues need to be addressed: 1) identification of rain and no-rain periods; and 2) determination of attenuation baseline. To solve these problems, this article adopts several machine learning algorithms based on the analysis of earth-space link signal characteristics. For the first issue, we choose the support vector machine as a classifier and the adaptive synthetic sampling algorithm is deployed to eliminate the effects caused by the data imbalance. For the second issue, the long short-term neural network is selected for the determination of attenuation baseline since it has a good ability to solve time-series problem. In terms of the rainfall inversion, we establish a new model by combining the back-propagation (BP) network and genetic algorithm (GA). The PL model is also used as a comparison. To validate the proposed method, we set up an earth-space link that receives the signal from AsiaSat5 in 12.32 GHz. The results demonstrate that the two issues are successfully addressed and the inversion of precipitation is also carried out. Compared to disdrometer, the correlation and mean absolute error of GA-BP model are 0.83 and 1.30 mm/h, respectively, indicating a great potential to use densely OELs for global precipitation monitoring.

Long-Short Term Neural Network Analysis of Center of Pressure of Gait

Detection of vascular dementia in early stages of cognitive impairment is difficult to do in a clinical setting since the earliest changes are often discrete and physiological in nature. One major aspect of this is gait patterns. This project utilizes force-sensing platforms, motion capture, and EMG sensors to unobtrusively collect biometric data from an individual's walking gait patterns. The data parameters gathered were center of pressure, gait phase and end of unloading/toe-ff events. By quantifying and analyzing machine learning algorithms, specifically deep learning time-series based models, onset patterns of vascular dementia are explored with an overarching goal of creating a system that will assist in understanding and diagnosing cases of vascular dementia. The proposed system provides a tool for which gait can be analyzed and compared over a long period of time and opens opportunity to increased personalization in health monitoring and disease diagnosis and provides an avenue to increase patient-centricity of medical care. Since gait is one of the early predictors of vascular dementia, we developed a long short-term neural network to predict the gait variations from which we can predict the onset of vascular dementia.

Remaining useful lifetime prediction via deep domain adaptation

In Prognostics and Health Management (PHM) sufficient prior observed degradation data is usually critical for Remaining Useful Lifetime (RUL) prediction. Most previous data-driven methods assume that training (source) and testing (target) condition monitoring data have similar distributions. However, due to different operating conditions, fault modes and noise, distribution and feature shift exist across different domains. This shift reduces the performance of predictive models when no target observed run-to-failure data is available. To address this issue, this paper proposes a new data-driven approach for domain adaptation in prognostics using Long Short-Term Neural Networks (LSTM). We use a Domain Adversarial Neural Network (DANN) approach to adapt remaining useful life estimates to a target domain containing only sensor information. We analyse our approach using the NASA Commercial Modular Aero-Propulsion System Simulation (C-MAPPS). The results show that the proposed method can provide more reliable RUL predictions than models trained only on source data for varying operating conditions and fault modes.

Spam detection in social media using convolutional and long short term memory neural network

As the use of the Internet is increasing, people are connected virtually using social media platforms such as text messages, Facebook, Twitter, etc. This has led to increase in the spread of unsolicited messages known as spam which is used for marketing, collecting personal information, or just to offend the people. Therefore, it is crucial to have a strong spam detection architecture that could prevent these types of messages. Spam detection in noisy platform such as Twitter is still a problem due to short text and high variability in the language used in social media. In this paper, we propose a novel deep learning architecture based on Convolutional Neural Network (CNN) and Long Short Term Neural Network (LSTM). The model is supported by introducing the semantic information in representation of the words with the help of knowledge-bases such as WordNet and ConceptNet. Use of these knowledge-bases improves the performance by providing better semantic vector representation of testing words which earlier were having random value due to not seen in the training. Proposed Experimental results on two benchmark datasets show the effectiveness of the proposed approach with respect to the accuracy and F1-score.

Long short-term memory neural network for traffic speed prediction using remote microwave sensor data

Neural networks have been extensively applied to short-term traffic prediction in the past years. This study proposes a novel architecture of neural networks, Long Short-Term Neural Network (LSTM NN), to capture nonlinear traffic dynamic in an effective manner. The LSTM NN can overcome the issue of back-propagated error decay through memory blocks, and thus exhibits the superior capability for time series prediction with long temporal dependency. In addition, the LSTM NN can automatically determine the optimal time lags. To validate the effectiveness of LSTM NN, travel speed data from traffic microwave detectors in Beijing are used for model training and testing. A comparison with different topologies of dynamic neural networks as well as other prevailing parametric and nonparametric algorithms suggests that LSTM NN can achieve the best prediction performance in terms of both accuracy and stability.

A Short-Term Neural Network Memory

Neural network memories with storage prescriptions based on Hebb’s rule are known to collapse as more words are stored. By requiring that the most recently stored word be remembered precisely, a new simple short-term neural network memory is obtained and its steady state capacity analyzed and simulated. Comparisons are drawn with Hopfield’s method, the delta method of Widrow and Hoff, and the revised marginalise model of Mezard, Nadal, and Toulouse.