Neural networks in building energy consumption prediction using pseudo dynamic approach

Building energy consumption prediction has been a major concern in the recent years. The energy consumption in building is essential for building manager to manage the building energy management system (BEMS) and energy operator in making energy decision services. This paper proposes a building energy consumption prediction with novel pseudo dynamic transitional methods with occupancy profiles and operational power characteristics of buildings. The case study is applied to heating energy consumption of buildings and results show that pseudo dynamic model have coefficient of correlation of energy consumption of error of 0.02%. Further, orthogonal array design is applied to the pseudo dynamic model to check the schedule of occupancy profile and operational power level of building energy consumption.