Optimizing Neural Network Hyperparameters with Swarm Intelligences for Commercial Buildings Load Classification

Akshay Gupta, Rahul Singhal, Rajesh Kumar

Research output: Chapter in Book/Report/Conference proceedingConference contributionpeer-review

1 Citation (Scopus)

Abstract

Advanced metering systems are crucial in smart grids for analyzing collected data with high accuracy, efficient control, and planning of energy demand and supplies. Therefore, an adaptable automated process is essential for classifying different types of building loads for the smart grid to comprehend their consumption pattern. In this paper, a method is suggested for the load classification of commercial buildings to improve classification accuracy with swift execution and less computational burden, unlike deep learning techniques. Moreover, using a neural network with the optimally selected hyperparameter leads to remarkable performance as the neural network's performance depends heavily on its hyperparameters.

Original languageEnglish
Title of host publicationProceedings of the 2021 IEEE 18th India Council International Conference, INDICON 2021
PublisherInstitute of Electrical and Electronics Engineers Inc.
ISBN (Electronic)9781665441759
DOIs
Publication statusPublished - 2021
Externally publishedYes
Event18th IEEE India Council International Conference, INDICON 2021 - Guwahati, India
Duration: 19 Dec 202121 Dec 2021

Publication series

NameProceedings of the 2021 IEEE 18th India Council International Conference, INDICON 2021

Conference

Conference18th IEEE India Council International Conference, INDICON 2021
Country/TerritoryIndia
CityGuwahati
Period19/12/2121/12/21

Keywords

  • Commercial building
  • Hyperparameter tuning
  • Load Classification
  • Load profiling
  • Particle search optimization
  • Random search

ASJC Scopus subject areas

  • Artificial Intelligence
  • Computer Science Applications

Fingerprint

Dive into the research topics of 'Optimizing Neural Network Hyperparameters with Swarm Intelligences for Commercial Buildings Load Classification'. Together they form a unique fingerprint.

Cite this