Adapting Internet of Things and Neural Network in Modelling Demand Side Energy Consumption and Management

Authors : Daniel ABODE, Omowumi OLASUNKANMİ, Waliu O APENA, Samson A OYETUNJI

Pages : 118-127

View : 33 | Download : 11

Publication Date : 2020-11-01

Article Type : Research Paper

Abstract :With the recent application of micro-grid system and off-grid renewable energy power system using internet of things insert ignore into journalissuearticles values(IoT); for the efficacy in demand side consumption management. The study employed usage of IoT supported with statistical initiative insert ignore into journalissuearticles values(logistic regression); to develop a knowledge-based solution for energy demand side consumption management. The research adopted two approaches to model the energy consumption pattern of a user with designed sensor nodes for environmental data acquisition insert ignore into journalissuearticles values(DAC); monitoring and state of switches insert ignore into journalissuearticles values(load points);. Leveraging on Internet of Things, the sensor node network transferred synchronized the data collected to Google Firebase cloud storage in real time. The data collected were used to train a logistic regression model for the prediction states of the receptacles and sensor readings. The study further investigated power usage insert ignore into journalissuearticles values(user); against human presence and hour insert ignore into journalissuearticles values(period); of the day separately and a mathematical model of the relationship was developed. The results revealed customer’s energy consumption; this includes models for the future projection. The model can be deployed to predict energy management on the demand side efficiency and availability indices. The models could support energy management including receptacle automation prediction and wastage monitoring.
Keywords : Electrical Energy, Machine Learning, MATLAB, Sensor Node, DAC

ORIGINAL ARTICLE URL