LSTM and Edge Computing for Big Data Feature Recognition of Industrial Electrical Equipment

With the rapid development of Industrial Internet of Things, the category and quantity of industrial equipment will increase gradually. For centralized monitoring and management of numerous and multivariate equipment in the intelligent manufacturing process, the equipment categories shall be identif...

Full description

Saved in:
Bibliographic Details
Published in:IEEE transactions on industrial informatics 2019-04, Vol.15 (4), p.2469-2477
Main Authors: Lai, Chin-Feng, Chien, Wei-Che, Yang, Laurence T., Qiang, Weizhong
Format: Article
Language:English
Subjects:
Appliance identification
Artificial neural networks
Big Data
Computational modeling
Data management
Edge computing
Electric equipment
Equipment costs
Feature extraction
Feature recognition
Industrial applications
Industrial development
Industrial equipment
Intelligent manufacturing systems
long-short term memory
Monitoring
nonintrusive load monitoring
smart grid
Smart meters
Training
Citations: Items that this one cites
Items that cite this one
Online Access:Get full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:With the rapid development of Industrial Internet of Things, the category and quantity of industrial equipment will increase gradually. For centralized monitoring and management of numerous and multivariate equipment in the intelligent manufacturing process, the equipment categories shall be identified first. However, manual labeling of electrical equipment needs high costs. For the purpose of recognizing industrial equipment accurately in manufacturing systems, this study adopts the long short-term memory to analyze big data features and build a nonintrusive load monitoring system. Edge computing is used to implement parallel computing to improve the efficiency of equipment identification. Considering the practical popularity, the fairly priced low-frequency Smart Meter is used to collect the appliance data. According to the proposed optimal adjustment strategy of parameter model, the average random recognition rate can achieve 88% and the average recognition rate of the continuous data of a single electrical equipment can achieve 83.6%.
ISSN:1551-3203
1941-0050
DOI:10.1109/TII.2019.2892818