An Application of Machine Learning to Estimate and Evaluate the Energy Consumption in an Office Room

There are no exact criteria for the architecture of openings and windows in office buildings order to optimize energy consumption. Due physical limitations this renewable source lack conscious control over its capabilities, amount light entering offices role daylight as a determined by how they constructed. In study, standard room dimensions, which suitable three five employees, compared computer simulations. DesignBuilder EnergyPlus utilized simulate office’s lighting This study presents new method estimating conventional consumption based on gene expression programming (GEP). A gravitational search algorithm (GSA) is implemented model results. Using input output data collected from simulation use, law underlying problem relationship between inputs outputs identified. has advantages being quick accurate, with required. Based effective parameters sensitivity analysis, four models evaluated. These used evaluate performance trained network statistical indicators. Among all GEP tested one lowest MAE (0.1812) RMSE (0.09146) highest correlation coefficient (0.90825) found be most accurate.