Energy Smart Meter operation improved by Machine Learning’s Decision-Support System and Internet Of Things

Gomathi.S, Meenakshi Sundararajan Engineering College (MSEC) ; Suruthi.R.S ,Meenakshi Sundararajan Engineering College (MSEC) ; Yamuna.S ,Meenakshi Sundararajan Engineering College (MSEC)

ESM, BN, NB, DT, Gini Index, RF, IOT

The electricity has become a part of daily life, which plays an important role in our homes and industries. The system is now focused on the growing demand of power and the need of finding the alternative energy source. The idea of a ‘smart city’ is the key solution to these power related problems, giving us a futuristic scope. Better understanding of domestic and commercial energy usage brings with it a problem of managing and classifying the sheer amount of data that comes along with it. The work proposal is basically to overcome the demand of power using smart meter in electric power consumption benefiting customer to monitor and manage the electric power usage. This idea is made easier by applying Machine Learning’s. Decision Support System an application of Artificial Intelligence (AI) to classify and distribute energy while managing and enhancing the other supporting features of an Electric S mart Meter (ES M) using Internet of Things (IOT). We plan on introducing smart meters as a ‘live’ communication tool connecting the provider with its customers, which will cause the electrical network industry to face a 360 degree turn around towards a customer-centric business. The system employs the Bayesian Network (BN) prediction model with the three machine learning model that is Naïve Bayes (NB), Decision Tree (DT) and Random Forest (RT). The ES M systems network model is based on the four cornerstones fundamental to IOT: sensing, computing, communication, and actuation.
    [1] Joseph Siryani, Ph.D. Candidate, Bereket Tanju, Ph.D., and Timothy Eveleigh, D.Sc [2] (2017) -A Machine Learning Decision Support System Improves the Internet of Things’ Smart Meter Operations. [3] http://www.bu.edu/sph/files/2014/05/bayesian-networks-final.pdf [4] http://chemeng.utoronto.ca/~datamining/dm c/decision_tree.htm [5] http://dni-institute.in/blogs/cart-decision-tree-gini-index-exp lained/ [6] http://www.tutorialspoint.com/r/r_random_f orest.htm [7] http://www.cloudbus.org/papers/Internet-of-Things -Vision-Future2012.pdf
Paper ID: GRDCF006002
Published in: Conference : National Conference on Advancement in Emerging Technologies (NCAET - 2018)
Page(s): 5 - 8