kumoh national institute of technology
Networked Systems Lab.

Love Allen Chijioke Ahakonye, Cosmas Ifeanyi Nwakanma, Jae Min Lee, Dong-Seong Kim, Time-Efficient Deep Learning-Based Energy Consumption Prediction for Smart Factory. The 26th International Conference on Emerging Technologies and Factory Automation (ETFA 2021), Vasteras, Sweden, 07-10 September, (N8 and N12) (S)
By :
Date : 2021-05-03
Views : 57

Energy consumption forecasting and planning in the smart factory is a vital aspect in the optimistic smart grid, by which energy loads require real-time forecast and organizing to subsist especially the fully combined difference between energy request and value. In this work, a convolutional neural network for non-image classification and forecasting with an extensive achievement appraisal of other algorithms such as Long-Short Term Memory (LSTM) and Bidirectional LSTM (Bi-LSTM)were compared. Focusing on the real-life application augmented smart grid stability dataset, the result based on computational complexity shows that Convolutional Neural Network (CNN) outperformed LSTM and Bi-LSTM which are popularly used. The dataset was preprocessed then split into 70%, 30% for training, testing respectively. Each models execution is analyzed on the basis of accuracy and computation time. The trial reveals 1D Convolutional Neural Networks to perform better in computation time compared to LSTM and Bi-LSTM, with an accuracy of 99.96%
(Total:16 articles / page:1/2 )