To search, Click below search items.


All Published Papers Search Service


Prediction of Energy Consumption from Renewable Solar Resources in a Smart Home using Markov Chain


J. Jasmine Christina Magdalene and B.S.E.Zoraida


Vol. 22  No. 9  pp. 433-438


Everything in the modern era is designed to be smart and a home is not an exception. A smart home is a home where all the devices are smart devices and are connected to a central point of access. Home automation expects energy efficiency. Energy generation and energy consumption are the momentous two sides of energy management. Hence prediction of energy usage from renewable resource becomes a dynamic option. In this paper the Markov Chain principle is used to forecast the amount of energy used from the solar energy generators across seasons. Markov Chain proves apt in handling random variables and it does not brood on historical data. It considers the present data alone to forecast the future. The dataset for this work is taken from Austin, Texas. A single home data is taken and the various seasons in Austin are considered as the state space. The energy consumption from solar is segregated based on the various seasons in Austin. Using these seasons as space state, a transition matrix is built and the future data is simulated. After the simulation of future data, Markov Chain is used to predict the amount of energy used from the solar generators during various seasons in the forth-coming year. Here forecasting is done for six years (2014 -2019) and the RMSE is calculated for each year. The loglikelihood is taken as a measure to prove that Markov Chain principle gives higher value than Bayesian and Markov Bootstrap therefore a better option.


Smart Home, Energy Efficiency, Transition Matrix, Markov Chain.