188 Hits in 5.8 sec

Cross-Domain Energy Consumption Prediction via ED-LSTM Networks

Ye TAO, Fang KONG, Wenjun JU, Hui LI, Ruichun HOU
2021 IEICE transactions on information and systems  
To take advantage of cross-domain resources, a hybrid energy consumption prediction framework is presented in this paper.  ...  This framework combines the long short-term memory model with an encoder-decoder unit (ED-LSTM) to perform sequence-to-sequence forecasting.  ...  This paper presents a short-term multistep prediction framework based on the ED and LSTM models for electricity consumption prediction, named ED-LSTM, that collectively considers cross-domain feature fusion  ... 
doi:10.1587/transinf.2020bdp0006 fatcat:evwu7crl5bakjju5dorfpvn5ai

Sequential Learning-Based Energy Consumption Prediction Model for Residential and Commercial Sectors

Ijaz Ul Haq, Amin Ullah, Samee Ullah Khan, Noman Khan, Mi Young Lee, Seungmin Rho, Sung Wook Baik
2021 Mathematics  
The proposed energy consumption prediction model outperforms existing models over publicly available datasets, including Household and Korean commercial building datasets.  ...  In this paper, we present a comparative analysis of a variety of deep features with several sequential learning models to select the optimized hybrid architecture for energy consumption prediction.  ...  [23] combined CNN with LSTM and presented a hybrid CNN-LSTM neural network approach for energy prediction with a very small RMSE value.  ... 
doi:10.3390/math9060605 fatcat:r6swl6ii45exffrbsi3blsa6vi

A Review on Deep Sequential Models for Forecasting Time Series Data

Dozdar Mahdi Ahmed, Masoud Muhammed Hassan, Ramadhan J. Mstafa, Aniello Minutolo
2022 Applied Computational Intelligence and Soft Computing  
Three deep sequential models, namely, artificial neural network (ANN), long short-term memory (LSTM), and temporal-conventional neural network (TCNN) along with their applications for forecasting time  ...  We conclude that the LSTM model is widely employed, particularly in the form of a hybrid model, in which the most accurate predictions are made when the shape of hybrids is used as the model.  ...  Experimental results showed that the DIDR-LSTM achieved a superior cross-domain RUL prediction accuracy [89] . Chen et al.  ... 
doi:10.1155/2022/6596397 fatcat:n6ufhsi7nza25hpj77csn7hrhu

Energy Consumption Forecasting in Korea Using Machine Learning Algorithms

Sun-Youn Shin, Han-Gyun Woo
2022 Energies  
In predicting energy consumption, classic econometric and statistical models are used to forecast energy consumption.  ...  In current literature, there are minimal comparison studies reviewing machine learning algorithms to predict energy consumption in Korea.  ...  Crompton and Wu [8] attempted at predicting the energy consumption of China via a Bayesian vector-based autoregression method.  ... 
doi:10.3390/en15134880 fatcat:qqqj4t6apjgitgpnf55tqn7mie

Smart Metering System Capable of Anomaly Detection by Bi-directional LSTM Autoencoder [article]

Sangkeum Lee, Hojun Jin, Sarvar Hussain Nengroo, Yoonmee Doh, Chungho Lee, Taewook Heo, Dongsoo Har
2021 arXiv   pre-print
Anomaly detection method based on the BiLSTM autoencoder is tested with the metering data corresponding to 4 types of energy sources electricity/water/heating/hot water collected from 985 households.  ...  The LSTM as online portals, allowing clients to access their profiles, for network can accomplish better characterization of time-series example, check their energy consumption trends.  ...  1 or the initial network represents training two uni-directional LSTM hidden state at time 0, and , , , represent the input, networks rather than one LSTM network on the input sequence.  ... 
arXiv:2112.03275v1 fatcat:n3pikubqhndw7gyg6d25dxnsdy

Predicting LoRaWAN Behavior: How Machine Learning Can Help

Francesca Cuomo, Domenico Garlisi, Alessio Martino, Antonio Martino
2020 Computers  
In this challenging case, the prediction of the inter-arrival time of packets has an error of about 3.5% for 77% of real sequence cases.  ...  Our study shows how profiling techniques enable a machine learning prediction algorithm even when training is not possible because of high error rates perceived by some devices.  ...  We thank Marco Zanghieri for assisting us with the study of the prediction activities with LSTM algorithm.  ... 
doi:10.3390/computers9030060 fatcat:zcvegz5wgndcnkch7rusdbukv4

Knowledge distillation with error-correcting transfer learning for wind power prediction [article]

Hao Chen
2022 arXiv   pre-print
This advantage also exists in terms of wind energy physics and computing efficiency, which are verified by the prediction quality rate and calculation time.  ...  This framework is the first time to incorporate knowledge distillation into energy forecasting, enabling accurate and economical constructions of turbine models by learning knowledge from the well-established  ...  And thanks to the support of UiT Arctic Centre for Sustainable Energy.  ... 
arXiv:2204.00649v1 fatcat:ydhawkvyxngkjhfpnp6uuhczhq

Privacy-aware Resource sharing in Cross-device Federated Model Training for Collaborative Predictive Maintenance

Sourabh Bharti, Alan McGibney
2021 IEEE Access  
His research interests include predictive maintenance, wireless networked systems, and the Internet of Things (IoT).  ...  INDEX TERMS SplitNN, federated learning, predictive maintenance, Industry 4.0.  ...  The deployment strategy is determined by the ED specific local parameters such as residual energy, workload etc.  ... 
doi:10.1109/access.2021.3108839 fatcat:qadewavoqjflbaf2huh3ymiwfu

Collaborative Learning on the Edges: A Case Study on Connected Vehicles

Sidi Lu, Yongtao Yao, Weisong Shi
2019 USENIX Workshop on Hot Topics in Edge Computing  
Our approach is built on top of the federated learning algorithm and long shortterm memory networks, and it demonstrates the effectiveness of driver personalization, privacy serving, latency reduction  ...  We choose the failure of EV battery and associated accessories as our case study to show how the CLONE solution can accurately predict failures to ensure sustainable and reliable driving in a collaborative  ...  memory networks (LSTMs) to predict failures, and we find LSTMs outper- form other methods based on our dataset. • We propose CLONE, a collaborative learning setting on the edges for connected vehicles  ... 
dblp:conf/hotedge/LuYS19 fatcat:jji6fszkp5dlzbdvwtf2o22l2u

Coarse Return Prediction in a Cement Industry's Closed Grinding Circuit System through a Fully Connected Deep Neural Network (FCDNN) Model

Morad Danishvar, Sebelan Danishvar, Francisco Souza, Pedro Sousa, Alireza Mousavi
2021 Applied Sciences  
Milling operations in various production processes are among the most important factors in determining the quality, stability, and consumption of energy.  ...  A combined event modeling for feature extraction and the fully connected deep neural network model to predict the coarseness of cement particles is proposed.  ...  Minimizing energy consumption while simultaneously improving product quality and process efficiency would be a major contribution to the cement industry, and reduce global energy consumption and greenhouse  ... 
doi:10.3390/app11041361 fatcat:lnt577js3bdo5p6pqqycp2hj5q

Deep Learning in Mobile and Wireless Networking: A Survey [article]

Chaoyun Zhang, Paul Patras, Hamed Haddadi
2019 arXiv   pre-print
Subsequently, we provide an encyclopedic review of mobile and wireless networking research based on deep learning, which we categorize by different domains.  ...  We first briefly introduce essential background and state-of-the-art in deep learning techniques with potential applications to networking.  ...  Liu et al. apply deep Q learning to reduce the energy consumption in cellular networks [389] .  ... 
arXiv:1803.04311v3 fatcat:awuvyviarvbr5kd5ilqndpfsde

ORCA: Enabling an Owner-centric and Data-driven Management Paradigm for Future Heterogeneous Edge-IoT Systems [article]

Jianli Pan, Jianyu Wang, Ismail AlQerm, Yuanni Liu, Zhicheng Yang
2021 arXiv   pre-print
owner-centric management paradigm named "ORCA" to address the gaps left by the owner-centric paradigm and empower the IoT assets owners to effectively identify and mitigate potential issues in their own network  ...  series behavior modeling, and integrate LSTM [13] with ED networks (LSTM-ED) for time-series behavior modeling.  ...  In the LSTM-ED model, we employ LSTM neurons in the hidden layers for both the encoder and the decoder networks to learn the time 2) "Synthesizing": Owner-centric Group and Subsystem Behavior Modeling:  ... 
arXiv:2102.03640v1 fatcat:s6tsjdjakjcqviaoo7y6noazim

Biometric User Identification Based on Human Activity Recognition Using Wearable Sensors: An Experiment Using Deep Learning Models

Sakorn Mekruksavanich, Anuchit Jitpattanakul
2021 Electronics  
The results for the two basic models, namely, the Convolutional Neural Network (CNN) and the Long Short-Term Memory (LSTM) deep learning, showed that the highest accuracy for all users was 91.77% and 92.43%  ...  Following this, the application of cross-correlation was conducted for the measurement of similarity, which indicated a 6.4% EER (Energy Efficiency Ratio).  ...  Human Activity Recognition via Machine Learning and Deep Learning The time series classification tasks are the main challenges in using HAR, which is when the person's movements are predicted by the use  ... 
doi:10.3390/electronics10030308 fatcat:khcvygvxc5hhjdajthgpnj7vd4

Automatic Gaze Analysis: A Survey of Deep Learning based Approaches [article]

Shreya Ghosh, Abhinav Dhall, Munawar Hayat, Jarrod Knibbe, Qiang Ji
2022 arXiv   pre-print
We analyze recent gaze estimation and segmentation methods, especially in the unsupervised and weakly supervised domain, based on their advantages and reported evaluation metrics.  ...  We conclude by discussing future research directions for designing a real-world gaze analysis system that can propagate to other domains including Computer Vision, Augmented Reality (AR), Virtual Reality  ...  Pinball LSTM. Similarly to encode contextual information along temporal domain, pinball LSTM [90] is proposed.  ... 
arXiv:2108.05479v3 fatcat:6qhwjojyqbdctjcwnjerflvyzi

Detecting Submerged Objects Using Active Acoustics and Deep Neural Networks: a Test Case for Pelagic Fish

Alberto Testolin, Dror Kipnis, Roee Diamant
2021 Zenodo  
To allow for real-time detection, we use a convolutional neural network, which provides the simultaneous labeling of a large buffer of signal samples.  ...  However, training the network directly on the real reflections with data augmentation techniques allowed to reach a more favorable precision-recall trade-off, approaching an ideal detection bound.  ...  The loss function to be minimized is the cross-entropy between the correct class (ground truth labels) and the network prediction.  ... 
doi:10.5281/zenodo.4983077 fatcat:bcfqvw5v6nc2bp2ddimfb423zu
« Previous Showing results 1 — 15 out of 188 results