The week's pick
Random Articles
Reseach Article
LSTM Deep Recurrent Neural Network Model for Voltage Abnormality Detection at IoT Gateway
| International Journal of Computer Applications |
| Foundation of Computer Science (FCS), NY, USA |
| Volume 177 - Number 9 |
| Year of Publication: 2019 |
| Authors: Jihad Qaddour, Navid Rajabi |
10.5120/ijca2019919442
|
Jihad Qaddour, Navid Rajabi . LSTM Deep Recurrent Neural Network Model for Voltage Abnormality Detection at IoT Gateway. International Journal of Computer Applications. 177, 9 ( Oct 2019), 9-13. DOI=10.5120/ijca2019919442
Abstract
In this paper, we have trained and tuned a Deep Neural Network (DNN) model to extract the household’s average voltage pattern and utilize it for predicting any potential abnormalities using Artificial Intelligence (AI). To be more specific, this model is a Recurrent Neural Network (RNN) that leverages the Long Short-Term Memory (LSTM) cell as a long-term dependency storage. The logic behind this architecture is to enhance the performance as well as solving the vanishing (and exploding) gradient problem of RNNs. The main objective of this project is to detect abnormalities at the edge (Internet of Thing (IoT) gateway) as a security-oriented proactive approach. Therefore, after training the model with a large-enough train part of the dataset and optimizing it after running the test part of the dataset, the model can be placed at the IoT gateway and recognize the future abnormalities in Time-series voltage dataset of any given household to avoid any potential malicious activity or damage.
References
- Understanding LSTM Networks, Recurrent Neural Networks, Posted on August 27, 2015, http://colah.github.io/posts/2015-08-Understanding-LSTMs/.
- Stanfordgroup,RrajPurkar, Jul2,2018,https://stanfordmlgroup.github.io/competitions/mura/
- J.Ortiz,“HouseholdPowerConsumption,”https://data.world/databeats/household-power-consumption.
- G. Hebrail, “Individual household electric power consumption Data Set,” Aug 30, 2012, https://archive.ics.uci.edu/ml/datasets/individual+household+electric+power+consumption.
- J. L. Reyes-Ortiz, D. Anguita, A. Ghio, L. Oneto, and X. Parra , “Human Activity Recognition Using Smartphones Data,April2013,Set”https://books.google.com/books?id=MtkoDwAAQBAJ&lpg=PP1&dq=Smart%20Cities%3A%20Foundations%2C%20Principles%2C%20and%20Applications&pg=PP1#v=onepage&q=Smart%20Cities:%20Foundations,%20Principles,%20and%20Applications&f=false.
- Googlebook,https://books.google.com/books?id=qEkzDwAAQBAJ&lpg=PP1&pg=PP1#v=onepage&q&f=false.
- LSTM_Networks/LSTM Demo.ipynb, “Generating Text using an LSTM Network (No libraries),” Aug 7, 2017, https://github.com/llSourcell/LSTM_Networks/blob/master/LSTM%20Demo.ipynb.
- T. Mitchell, A. Singh, “Machine Learning,” Carnegie MellonUniversity,2011,http://www.cs.cmu.edu/~aarti/Class/10601/slides/Nnets_11_3_2011.pdf.
- S. D. Ali, “The Evolution and Core Concepts of Deep Learning & Neural Networks,” August 2016,https://www.analyticsvidhya.com/blog/2016/08/evolution-core-concepts-deep-learning-neural-networks/.
- LSTM prediction using Neural Networks, Jan 25, 2018,https://github.com/jaungiers/LSTM-Neural-Network-for-Time-Series-Prediction.
Index Terms
Keywords