Main Article Content
The demand for the power distribution to the residential buildings is a non consistent form that depends on the inhabitant requirement, occupancy dynamism and the appliances that were working in it. The optimization of power consumption and various different building parameters tends to the reduced power consumption and maximize the efficiency of power management integrating with the forecasting of energy demands. The occupancy dynamism in the residential area possess random pattern of energy utilization and needs an efficient optimization algorithm for the maintenance of occupancy centric energy consumption. With an aim of addressing this need, this paper proposes a novel model of Chess Optimization (CO) algorithm with three phase namely preprocessing, feature extraction and classification. The proposed algorithm utilizes the datasets effectively for the comparison of trained and testing phase to fix the global active power and time series data set values. The performance analysis of the proposed chess optimization algorithm is done at multiple levels of global active with time and smart meters with respect to time. The dataset related to the hourly energy consumption has been effectively used for prediction of high power and low power consumption meters in which the appliances were classified under various levels of smart meters. The experimental results proves that the proposed optimization algorithm possess minimal energy difference (with a standard deviation of 0.1) between the user set parameters and the real time measured parameters, also proves to be an effective energy utilization management with a minimal power consumption.