Applicability of Multivariate Linear Regression in Building Energy Demand Estimation

The vision of the research project is to find an energy optimal building configuration, suitable for specified requirements and restrictions. first step on this way create a measure compare configurations, faster than explicit energetic simulations. current study examines applicability multivariate linear regression support solution optimization problems. During study, models were created estimate expected annual heating demand configurations examined their accuracy Between examinations, modified so that complexity was increased only such extent approximation still sufficiently accurate. result model estimated output unknown descriptive variables with 0% relative error 1.6% standard deviation. R2 point estimates 0.9884. Based these, considered applicable in search space defined by training patterns.